[go: up one dir, main page]

WO2018122819A1 - 一种基于双目图像分析的沥青路面病害检测系统 - Google Patents

一种基于双目图像分析的沥青路面病害检测系统 Download PDF

Info

Publication number
WO2018122819A1
WO2018122819A1 PCT/IB2017/058549 IB2017058549W WO2018122819A1 WO 2018122819 A1 WO2018122819 A1 WO 2018122819A1 IB 2017058549 W IB2017058549 W IB 2017058549W WO 2018122819 A1 WO2018122819 A1 WO 2018122819A1
Authority
WO
WIPO (PCT)
Prior art keywords
crack
image
information
road surface
asphalt pavement
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/IB2017/058549
Other languages
English (en)
French (fr)
Inventor
杜豫川
张晓明
蒋盛川
仇越
常光照
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Intelligent Transportation Co Ltd
Tongji University
Original Assignee
Shanghai Intelligent Transportation Co Ltd
Tongji University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Intelligent Transportation Co Ltd, Tongji University filed Critical Shanghai Intelligent Transportation Co Ltd
Priority to CN201780056387.4A priority Critical patent/CN109716108B/zh
Priority to US16/474,710 priority patent/US11486548B2/en
Priority to GB1905912.0A priority patent/GB2571016B/en
Publication of WO2018122819A1 publication Critical patent/WO2018122819A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • FMECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F17STORING OR DISTRIBUTING GASES OR LIQUIDS
    • F17DPIPE-LINE SYSTEMS; PIPE-LINES
    • F17D5/00Protection or supervision of installations
    • F17D5/02Preventing, monitoring, or locating loss
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N3/00Investigating strength properties of solid materials by application of mechanical stress
    • G01N3/08Investigating strength properties of solid materials by application of mechanical stress by applying steady tensile or compressive forces
    • EFIXED CONSTRUCTIONS
    • E01CONSTRUCTION OF ROADS, RAILWAYS, OR BRIDGES
    • E01CCONSTRUCTION OF, OR SURFACES FOR, ROADS, SPORTS GROUNDS, OR THE LIKE; MACHINES OR AUXILIARY TOOLS FOR CONSTRUCTION OR REPAIR
    • E01C23/00Auxiliary devices or arrangements for constructing, repairing, reconditioning, or taking-up road or like surfaces
    • E01C23/01Devices or auxiliary means for setting-out or checking the configuration of new surfacing, e.g. templates, screed or reference line supports; Applications of apparatus for measuring, indicating, or recording the surface configuration of existing surfacing, e.g. profilographs
    • EFIXED CONSTRUCTIONS
    • E01CONSTRUCTION OF ROADS, RAILWAYS, OR BRIDGES
    • E01CCONSTRUCTION OF, OR SURFACES FOR, ROADS, SPORTS GROUNDS, OR THE LIKE; MACHINES OR AUXILIARY TOOLS FOR CONSTRUCTION OR REPAIR
    • E01C7/00Coherent pavings made in situ
    • E01C7/08Coherent pavings made in situ made of road-metal and binders
    • E01C7/18Coherent pavings made in situ made of road-metal and binders of road-metal and bituminous binders
    • FMECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F17STORING OR DISTRIBUTING GASES OR LIQUIDS
    • F17DPIPE-LINE SYSTEMS; PIPE-LINES
    • F17D5/00Protection or supervision of installations
    • F17D5/02Preventing, monitoring, or locating loss
    • F17D5/06Preventing, monitoring, or locating loss using electric or acoustic means
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01MTESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M3/00Investigating fluid-tightness of structures
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01MTESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M3/00Investigating fluid-tightness of structures
    • G01M3/002Investigating fluid-tightness of structures by using thermal means
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01MTESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M3/00Investigating fluid-tightness of structures
    • G01M3/38Investigating fluid-tightness of structures by using light
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01MTESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M5/00Investigating the elasticity of structures, e.g. deflection of bridges or air-craft wings
    • G01M5/0033Investigating the elasticity of structures, e.g. deflection of bridges or air-craft wings by determining damage, crack or wear
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01MTESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M5/00Investigating the elasticity of structures, e.g. deflection of bridges or air-craft wings
    • G01M5/0066Investigating the elasticity of structures, e.g. deflection of bridges or air-craft wings by exciting or detecting vibration or acceleration
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84Systems specially adapted for particular applications
    • G01N21/88Investigating the presence of flaws or contamination
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84Systems specially adapted for particular applications
    • G01N21/88Investigating the presence of flaws or contamination
    • G01N21/95Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/42Road-making materials
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84Systems specially adapted for particular applications
    • G01N21/88Investigating the presence of flaws or contamination
    • G01N21/8806Specially adapted optical and illumination features
    • G01N2021/8845Multiple wavelengths of illumination or detection
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N25/00Investigating or analyzing materials by the use of thermal means
    • G01N25/72Investigating presence of flaws

Definitions

  • Asphalt pavement disease detection system based on binocular image analysis
  • the present invention belongs to the field of intelligent transportation systems and information technology, and relates to five subsystems of using crack development degree detection model establishment, information collection, information analysis, information transmission and information release, including mobile platform, processing end, and temperature measuring infrared heat.
  • Image general image collector, vibration sensor, image processing technology, edge cloud computing technology, embedded system, system integration technology, real-time positioning technology to detect pavement crack disease information, including the number, type, location and development of cracks degree.
  • the gray scale information and temperature information of cracks in asphalt pavement can be obtained.
  • the gray scale information can realize the target of conventional image-based crack identification, and the temperature information can be used to detect the crack development degree, and the crack detection can be realized by using the above technology.
  • the system is built on the mobile platform to achieve fast and accurate detection of cracks in asphalt pavement.
  • BACKGROUND OF THE INVENTION With the development of road traffic, the problem of maintenance and management of road pavement has become more and more prominent, and the damage detection of road pavement has become one of the priorities of related road maintenance departments. At present, the road damage data obtained by China's pavement management system is still dominated by manual testing, which is inefficient and subject to subjective factors. With the rapid development of digital image processing technology in recent years, many researchers at home and abroad have conducted extensive research on image recognition pavement cracks and achieved good results.
  • the existing road surface disease detection vehicle adopts multi-dimensional comprehensive and sophisticated technology, the detection vehicle is expensive and the use cost is high, and it is difficult to achieve full coverage high-frequency inspection of all grades of roads, resulting in insufficient decision support data for intelligent maintenance, therefore, the road
  • the wisdom of the management is to be able to achieve high-frequency inspection equipment covering all levels of highways, in order to collect sufficient damage data to give decision support, and realize the intelligent management of the entire road network.
  • Zhao Chunxia and Tang Zhenmin of Nanjing University of Science and Technology successfully developed the N-1 road surface inspection vehicle.
  • the test vehicle can collect data such as surface crack image, flatness, and rut at 70km/h, and then process the data offline to generate a report.
  • the birth of the car promoted the development of automatic road detection vehicles in China.
  • the car uses a high-precision line-array camera to capture images of the road surface and uses laser lighting technology. Uninterrupted acquisition of the road image is achieved under the control of the encoder's synchronous trigger signal.
  • the system has a crack resolution of up to lmm, a maximum detection width of 4m and a top speed of 120km/h. current technology
  • Patent document CN200910222882.5 discloses an image-based pavement crack detection system, the system comprising: a control module, generating a trigger signal according to the received GPS positioning information, and outputting a trigger signal to the image to the acquisition module; the image acquisition module, according to The received trigger signal acquires the road surface and the road sign image to obtain the road surface image and the road sign image, and outputs the road surface image and the road sign image output value data processing module; the data processing module splices the first digital image to obtain a continuous digital image of the road surface, and utilizes The image recognition method identifies the crack in the continuous digital image of the road surface, obtains the first digital image including the crack, the GPS positioning information, and the second digital image, and determines the position information of the road surface crack according to the GPS positioning information and the second digital image, and outputs .
  • the invention also provides a theory of an image based pavement crack detection method. With the disclosed system and method, the accuracy and detection efficiency of the title detection can be achieved.
  • Prior art 2 provides a theory
  • Patent document CN201310252824.3 discloses a road surface crack detection system, which comprises an information acquisition system, a data storage display system, a power management system, a data processing system, and a GPS positioning system.
  • the information of the road surface is recorded by the information acquisition system, and the location of the disease is determined by combining the GPS and highway mileage stations of the GPS positioning system.
  • the digital image processing technology of the data storage display system is used to identify the disease, and the objective and stable detection result can be obtained. It can quickly and accurately locate and measure diseases.
  • the invention aims at the road crack detection work, realizes the real-time monitoring of related information, data storage and playback, real-time correlation between GPS information and mileage station number, and measurement of crack state information, and designs an on-board detection system to meet the application of actual pavement crack detection. Claim. Prior art 3
  • Patent document CN201410269998.5 discloses a road surface crack detecting device based on depth and gray image, comprising: a carrier platform (1); an imaging device (6), a line laser (7), a computing device (on the carrier platform) 8), wherein the carrier platform (1) is used to move along the road direction during the crack detection process, and the line laser (7) is used to illuminate the vertical road surface while the carrier platform (1) moves, and the camera device (6) ) used to move the laser beam reflected by the linear laser along the oblique direction at a diagonal angle, while the carrier platform (1) moves, and image a road section for each shot to generate laser lines of multiple road sections.
  • the computing device (8) is configured to generate depth data of the road section and gray scale data from the image data of the laser line of each road section, and splicing the depth data and the gray scale data of each road section Image data of a section of road for crack identification.
  • Degree of crack development The degree of development mentioned in the invention characterizes the extent of damage to the pavement caused by cracks in the asphalt pavement and the severity of recent damage. It covers the traditional crack severity classification, and the width, depth and crack of the crack. Humidity and the like may be related to factors that increase the severity of crack damage. These factors are included to characterize the level of development of cracks from the beginning to the present.
  • Classification function The temperature difference data is linearly classified according to the degree of crack development by the support vector machine. The development degree is divided into three levels of 1, 2, and 3, and 3 is the most serious. Then there will be one between 1, 2 and 2, 3.
  • the line is the dividing line, and this line expression is the classification function.
  • Crack zone An area of asphalt pavement that includes not only the fracture zone itself, but also a section of the road surface within a certain range, including the requirements of image processing and crack identification.
  • the temperature difference data obtained by taking the measured atmospheric temperature into the two classification functions of the crack development degree detection model.
  • Measured temperature difference data Temperature difference data between the crack area and the road surface area in the crack area obtained by image processing the acquired infrared image of the crack area.
  • Development level A number in 1, 2 or 3 whose size reflects the degree of development of the fracture. The larger the number, the more serious the degree of crack development.
  • Processing end Responsible for accepting the collected raw data, storing and real-time processing, and finally transmitting the processed data wirelessly to the remote server.
  • Support Vector Machine is a machine learning method based on statistical learning theory developed in the mid-1990s. It seeks to improve the generalization ability of learning machines by seeking structural risk minimization, and realizes the minimum of empirical risk and confidence range. In order to achieve a good statistical rule in the case of a small sample size.
  • support vector machines SVMs, which also support vector networks
  • SVMs are supervised learning models related to related learning algorithms, which can analyze data, identify patterns, and use for classification and regression analysis. Given a set of training samples, each marked as belonging to two categories, an SVM training algorithm builds a model, assigning new instances to one class or other classes, making it a non-probabilistic binary linear classification. Generally speaking, it is a two-class classification model.
  • the basic model is defined as the linear classifier with the largest interval in the feature space. That is, the learning strategy of the support vector machine is to maximize the interval and finally transform into a convex quadratic. Solving the problem of planning.
  • the present invention provides a road surface disease detection system based on infrared binocular image analysis and power spectral density.
  • the system includes five subsystems: development degree detection model establishment, information collection, information analysis, information transmission and information release.
  • the infrared thermal image obtained by the infrared camera can not only obtain the gray information of cracks and pavements, but also get their temperature information.
  • the detection method uses a crack development degree detection model based on infrared image analysis.
  • the acceleration change during the running of the vehicle is mainly caused by the change of the road elevation. Therefore, the road elevation change is regarded as the system excitation, the acceleration of the vehicle body is regarded as the system response, and the power level density and the linear time-invariant system theory can be used to derive the international leveling.
  • the conversion formula of the degree index (IRI) and the sprung acceleration power spectral density can improve the calculation accuracy by measuring the acceleration values of different positions in the car, and the vehicle speed correction coefficient can correct the influence of the vehicle speed change on the calculation result.
  • the influence of the bridgehead jump on the human comfort is calculated, and the impact coefficient and the system response method are used.
  • Estimate the equivalent impact force of the bridgehead is calculated, and the detection model of road surface flatness is indirectly estimated by measuring the Z-axis acceleration variation in the vehicle.
  • the method has the advantages of convenience, low consumption and economic rationality, and is suitable for a wide range of road surface flatness measurement.
  • using the mature mathematical methods such as wavelet theory, Kalman filter, linear time-invariant system, and the concept of annoying rate in experimental psychology, we can evaluate the phenomenon of bridgehead jumping.
  • IRI 0.782ax fe i + l .300 x ri ht - 3.442 .
  • the detection system of the pavement crack development degree based on infrared thermography is used to construct the detection system.
  • the actual detection system needs to use the infrared binocular camera to collect the road image.
  • the binocular camera consists of a common camera and an infrared camera, which simultaneously acquires the image of the asphalt pavement and processes it to identify the presence, absence, location, size and development of the crack.
  • the fusion of the two image data can improve the invention.
  • the robustness of the detection method is used to construct the detection system.
  • the crack detection system is modular in design and can be combined with other pavement disease detection modules.
  • the crack detection module includes a camera fixture, a binocular camera, a data transmission line, a vehicle terminal, a GPS receiver, and inertial navigation.
  • the camera fixture can be customized with a fixed iron frame to ensure that the binocular camera can be reliably fixed on the engineering vehicle.
  • the vehicle can be used to control the binocular camera angle while fixing the camera.
  • the pan/tilt has at least two degrees of freedom. It can be rotated in the horizontal and vertical directions respectively, and more degrees of freedom will bring higher operability of the gimbal.
  • the minimum configuration of the infrared camera is: 320 X 240 resolution, replaceable lens, maintenance-free uncooled microbolometer, microscopy and close-up measurement, data transfer speeds up to 60 Hz.
  • the normal image camera supports 1080P HD image real-time output.
  • the data transmission line includes at least two, which respectively support high-frequency transmission of high-definition ordinary images and infrared images, and supports 1080p video transmission of at least 100 Hz.
  • the vehicle terminal includes two schemes: one is front-end processing, the embedded PC is used for real-time video stream processing, and the data transmission is processed through the 3g ⁇ 4g network, and the embedded PC can be compatible with the data interface of the ordinary image and the infrared image.
  • the processor supports real-time processing of the video stream; the other is the back-end processing.
  • the vehicle-mounted terminal serves as the front end of the data acquisition, and is only responsible for data acquisition and storage. The configuration and development requirements are lower, and an interface is required for the third-party device to be accessed after the acquisition.
  • This module uses the video image stitching technology to restore the longitudinal image of the road surface during the entire acquisition process, and then separately cut along the length direction to identify the accuracy and stability of the asphalt pavement crack detection.
  • This system idea image processing algorithm requires high requirements and relatively few devices.
  • the GPS accepts the device and inertial navigation, which together position the acquisition device to ensure accuracy and real-time. The accuracy of positioning needs to be within 10m.
  • Another crack detection module includes a camera fixing device, a binocular camera, a data transmission line, a vehicle terminal, a GPS receiving device, an inertial navigation, a photoelectric encoder, and a synchronization controller (refer to CN104749187).
  • the photoelectric encoder is mounted on the wheel center axis of the vehicle moving platform for measuring the running speed and distance of the vehicle moving platform; the GPS receiver, Mounted on the vehicle mobile platform for high-precision positioning and timing of the vehicle-mounted mobile platform; the inertial navigation is installed on the vehicle-mounted mobile platform for the GPS receiver in the tunnel to receive the G PS signal Next, measuring the position and attitude data of the vehicle mobile platform to achieve high-precision position estimation inside the tunnel; the synchronization controller is installed on the vehicle mobile platform, and is used for synchronizing the image acquisition time of the ordinary camera and the infrared camera to ensure that both have A unified time and space benchmark. This method accurately measures the vehicle speed through the photoelectric encoder.
  • the synchronous controller automatically controls the binocular image acquisition time according to the vehicle speed and the field of view of the binocular camera, ensuring that the road images of two adjacent effective clips have good continuity.
  • the collected lanes are completely covered and do not overlap each other.
  • the front-end processing or storage function is performed. This system idea requires higher equipment and lower image processing algorithms.
  • the fracture development degree detection model establishes a subsystem:
  • the infrared image of the crack region of at least 10 samples of the asphalt pavement is collected, and the atmospheric temperature of the sample and the degree of crack development in the crack region of the sample are recorded, and the sample temperature difference data of the sampled sample is obtained after the image processing. Taking the atmospheric temperature as the abscissa, the temperature difference data of the crack region and the road surface region in the corresponding infrared image is plotted on the ordinate as the coordinate point.
  • the sample data required in the model establishment phase is shown in Figure 1.
  • the coordinate points are linearly classified using a support vector machine. Taking the atmospheric temperature as the abscissa, the crack and the road surface temperature difference as the ordinate to draw points, there are three levels of 1, 2, 3, the greater the number, the more serious the development degree, as shown in Figure 8, the classification function diagram, you can get the following two A classification function, as shown in equation (4) (5).
  • ⁇ 12 ⁇ 12 ⁇ +6 12 ( 5 )
  • ( °C ) is the atmospheric temperature
  • ⁇ CO is the temperature difference between the asphalt pavement and the crack
  • 12 is the linear classification function coefficient
  • the value range is 0.0075 ⁇ 0.0100
  • 2 linear classification function constant term the value range is 0.4 ⁇ 0.65.
  • the detection result is based on the following judgment to first calculate ⁇ 12 and ⁇ 23 according to the atmospheric thermometer, and then compare the measured ⁇ with ⁇ 12 and ⁇ 23
  • AT ⁇ AT U has a development degree of 1; ⁇ 12 ⁇ ⁇ ⁇ ⁇ 23 has a development degree of 2, and AT ⁇ AT 23 has a development degree of 3.
  • Now data acquisition is performed on other asphalt roads to verify the accuracy of the above detection model. From the above analysis, when the road surface is completely dry, the temperature difference between the asphalt pavement and the crack is mainly related to the temperature, and the temperature difference between the crack and the road surface is The degree of development of the crack is related. Therefore, the temperature difference between the crack and the road surface can be detected by an infrared camera, and then the above-mentioned classification function / ⁇ / 2 is used to detect the degree of crack development.
  • test environment meets the requirements, that is, the experimental environment described above: During the sunny day, between 8 am and 4 pm, the road surface is clean and completely dry. Then, the infrared thermal imager is used for data acquisition, and the temperature at the time of collecting each picture is recorded, and the temperature difference between the crack and the road surface is obtained after the treatment, and the development degree grading thresholds ⁇ 12 and ⁇ 23 are calculated according to the temperature. Table 2 Model verification results crack level actual quantity theoretical quantity relative error
  • the data acquisition layer includes a vehicle positioning device, an active infrared image sensor, a temperature measuring infrared camera, and a three-axis acceleration sensor.
  • Vehicle positioning equipment Due to its convenience and high measurement efficiency, the vehicle mobile measurement system has been applied in many projects such as urban planning, road inspection, digital city, etc., and has become an important direction of current surveying and mapping. In order to achieve high-precision in-vehicle mobile measurement, higher requirements are required for the global positioning of the Global Navigation Sat-elite System (GNSS).
  • GNSS Global Navigation Sat-elite System
  • the characteristics of the mobile measurement system determine that the on-board GNSS positioning requires high precision, high frequency and fast dynamic positioning.
  • BDS Bei Dou Navigation Satellite System
  • the vigorous revitalization of the Russian GLONASS system and the modernization of the US GPS system there will be more high-quality navigation satellites in the air.
  • Multi-system combined observations can greatly increase the number of observation satellites and significantly improve the geometric distribution of satellites.
  • the present invention adopts the latest Beidou or GPS-based positioning technology to realize accurate real-time refresh of vehicle position information, and the device can continuously transmit position information to the processing terminal in a certain frequency range, including time and latitude and longitude.
  • Infrared camera try to avoid direct light source, because the infrared light power control part is based on the photosensitive resistor installed on the infrared light board to control whether the working power of the infrared light is turned on or not. Infrared camera field should try to avoid all black objects, open spaces, water and other objects that absorb infrared light.
  • the infrared light of the CCD camera is reflected on the object by infrared light to form an image on the CCD camera lens. Being absorbed or weakened will greatly impair the effective illumination of the infrared lamp.
  • the active infrared image sensor device realizes the detection of grayscale image acquisition of the road surface, and the image quality is at least 720p, and the digital image is wired and transmitted to the data processing end.
  • thermal imaging cameras Until the 1960s, thermal imaging technology was used in non-military applications. Although early thermal imaging systems were cumbersome, slow in data acquisition, and poorly resolved, they were used in industrial applications such as inspection of large transmission and distribution systems. In the 1970s, the continued development of military applications led to the first portable system. The system can be used in applications such as building diagnostics and non-destructive testing of materials.
  • thermal imaging systems of the 1970s were rugged and very reliable, but their image quality was poor compared to modern thermal imaging cameras.
  • thermal imaging technology has been widely used in medical, mainstream, and building inspections. Once calibrated, the thermal imaging system can produce a complete image of the radiation so that the radiant temperature at any location in the image can be measured.
  • a radiation image is a thermal image that contains calculated values of temperature measurements at various points within the image.
  • the safe and reliable thermal imager cooler has been modified to replace the long-established compressed or liquefied gas used to cool the thermal imager.
  • a lower cost, duct-based thermoelectric light guide camera (PEV) thermal imaging system has been developed and mass produced. Although radiometric measurements are not possible, the PEV thermal imaging system is lightweight, easy to carry, and operates without cooling.
  • a focal plane array is an image sensing device consisting of an array (usually rectangular) of infrared sensing detectors located at the focal plane of the lens.
  • thermography used can be referred to the following but are not limited to:
  • Image frequency 50Hz (100/200Hz with window);
  • FPA Focal Plane Array
  • Object temperature range -20 ° C ⁇ 120 ° C;
  • Triaxial Accelerometers Triaxial accelerometers mostly use piezoresistive, piezoelectric, and capacitive operating principles. The resulting acceleration is proportional to changes in resistance, voltage, and capacitance, and is collected by corresponding amplification and filtering circuits. This and the ordinary acceleration sensor are based on the same principle, so in a certain technology, three single axes can be turned into a three-axis. For For most sensor applications, two-axis accelerometers are already available for most applications. However, some applications are concentrated in three-axis accelerometers such as digital mining equipment, valuable asset monitoring, collision monitoring, measuring building vibration, wind turbines, wind turbines and other sensitive large structural vibrations.
  • the invention adopts an integrated design acceleration sensor.
  • the sensor module has an infinite transmission module and a lithium battery, which can realize the vibration signal acquisition of the rear axle of the vehicle, and the frequency can be up to 200Hz.
  • the acceleration sensor is a collection device of the flatness detection data detection module
  • the two cameras are collection devices for road surface disease detection such as cracks
  • the positioning device is an auxiliary device for realizing automatic detection processing.
  • the acquisition equipment and auxiliary equipment of each module can be designed with reference to Figure 2 below, which can be fixed on various mobile platforms.
  • the vehicle mount includes two camera platforms and three mounts. Each mount has 2 to 4 magnet adsorption devices. The magnet mounts on the three brackets can be used to fix the two camera platforms. There are two screw holes at the same time, and it is necessary to use screws to fix the acquisition system to the vehicle equipment. The bracket needs to ensure the stability of the camera and the reliability in the rapid movement of the vehicle.
  • the data transmission layer includes two methods: wired transmission and wireless transmission.
  • the wired transmission includes two channels of video and positioning information.
  • the normal image uses a high-definition video transmission line, and the infrared image can be transmitted using a high-definition video transmission line or a network cable.
  • the transmission of the vibration signal adopts the wireless transmission mode. Referring to the zigbee or WIFI mode, the invention adopts the zigbee transmission mode, and the wifi is used as a technical reserve.
  • ZigBee is a highly reliable wireless data transmission network similar to CDMA and GSM networks.
  • the ZigBee data transmission module is similar to a mobile network base station.
  • the communication distance is from standard 75m to several hundred meters, several kilometers, and supports unlimited expansion.
  • ZigBee is a wireless data transmission network platform consisting of up to 65,000 wireless data transmission modules. Throughout the network, each ZigBee network data transmission module can communicate with each other.
  • the distance between each network node can be Standard 75m unlimited expansion.
  • the ZigBee network is mainly established for industrial field automation control data transmission. Therefore, it must be simple, easy to use, reliable in operation, and low in price.
  • the mobile communication network is mainly established for voice communication.
  • the value of each base station is generally more than one million yuan, and each ZigBee "base station" is less than 1,000 yuan.
  • Each ZigBee network node can not only be used as a monitoring object, for example, the sensor connected to it directly collects and monitors data, and can also automatically transfer data data transmitted by other network nodes.
  • each ZigBee Network Node can be wirelessly connected to multiple isolated child nodes (RFDs) that do not undertake network information relay tasks within the coverage of their own signals.
  • Data transmission is designed on the principle of reliability, real-time and stability to ensure efficient communication between the data processing layer and the data acquisition layer.
  • the command information of the control terminal needs to be sent to the acquisition device.
  • the network transmission is shown in Figure 3.
  • the data storage device has a built-in and processing end, and mainly stores system processing logs, original disease image data, original vibration data, and their corresponding spatiotemporal coordinates for later processing and large data accumulation.
  • the above is the data transmission from the acquisition layer to the processing layer, and the transmission from the processing layer to the server database.
  • the real-time transmission is only for the disease information after the data processing, and mainly uses the wireless network for transmission, including the 3g/4g network, and the data.
  • Narrowband IoT transmission technology is used under the conditions allowed.
  • This layer of data transmission is only one-way transmission from the processing end to the server center.
  • the data processing layer includes ordinary image processing, infrared image processing and vibration signal processing.
  • the image frame obtained by the image processing is matched with the position information, and finally the road pavement disease type and position are obtained by processing; the vibration signal processing, and the position information are further processed. Match, get the flatness along the road and the abnormal jumping information.
  • Image denoising is suitable for ordinary images and infrared images.
  • digital images are often affected by imaging equipment and external environmental noise during digitization and transmission. It is called a noisy image or a noisy image.
  • image denoising The process of reducing noise in digital images is called image denoising, sometimes referred to as image denoising.
  • Noise is an important cause of image interference.
  • An image may have a variety of noises in practical applications, which may be generated in transmission or may be generated in processing such as quantization.
  • the methods of image noise reduction mainly include the following types: mean filter, adaptive Wiener filter, median filter, morphological noise filter, wavelet denoising, etc.
  • Edge detection is a very common detection process in various image detection applications, because the features of interest usually have significant changes in the local grayscale of the image, which are quite different from the background image, and there are edges where these changes are strong.
  • the edge information of the image can be decomposed into edge direction and edge amplitude characteristics. Usually, the amplitude of the edge changes more gently along the edge direction, while the vertical edge changes sharply in the direction of the vertical edge. According to this characteristic of the edge, scholars have proposed a number of differential operators based on first or second order to achieve edge detection.
  • the invention uses the Canny operator for edge detection to identify crack diseases, which is the core algorithm for crack detection.
  • the Canny operator has the best detection effect. Its main feature is that the noise control is very good, the pseudo crack can be removed, the image is clean, and the morphological processing has great advantages.
  • the key parameters of the Canny operator are the value of Alpha and two high and low threshold parameters. The non-maximum suppression of the gradient amplitude reduces the splitting and joining of the pseudo-cracks and the double threshold, so that the edges are continuously integrated and clearly identifiable.
  • threshold segmentation The latter step in the extraction of the edge information is the threshold segmentation. If the appropriate threshold segmentation is not performed, the edge information will be very cluttered, and the pseudo-cracks occupy the entire image region, so the step of threshold segmentation is crucial.
  • the basic idea of threshold segmentation is based on the difference in the grayscale values between the target and the background in the original image. The image is divided into two parts: the target (foreground) and the background by setting one or more thresholds.
  • Threshold segmentation Two, global threshold methods, including bimodal method, Otsu method, minimum error method and one-dimensional maximum entropy method, etc., the present invention adopts a one-dimensional maximum entropy method.
  • the present invention employs a mathematical morphology approach to solve these problems.
  • Mathematical morphology is a discipline based on set theory. It is a tool for approximate mathematical analysis of images based on the form of images. The core idea is to measure and process the original image with a structured element with a certain shape. Mathematical morphology can simplify image data and can be applied to areas such as image segmentation, noise suppression, and image restoration.
  • the present invention utilizes a selection method based on the morphological characteristics of the fusion crack, and can achieve a correct rate of more than 90%.
  • Anisometry Ra /Rb ( 3)
  • the crack usually presents a relatively narrow area in the small image.
  • Each ellipse Region to be determined is circumscribed by an ellipse, and the Anisometry parameter of each Region in the region is obtained, and the maximum is recorded. Anisometry, as the basis for judgment.
  • the invention first introduces the idea of block detection of pavement image, and then performs edge detection on the crack image of the road surface.
  • we use an improved Canny operator to detect the crack edge of the road surface and the experimental results show the detection effect. More excellent.
  • the invention considers the applicability of the complex condition of the road surface, selects the one-dimensional maximum entropy method to segment the image, and makes a comparative experiment. Then it focuses on how to remove the cracked orphans that have been segmented.
  • the invention adopts mathematical morphology to propose the pseudo-cracks of the isolated points, and has achieved good results.
  • the invention describes a method for determining cracks, and the size of the crack is counted to meet the requirements of PCI statistics.
  • Infrared image processing is divided into two major steps. First, grayscale processing is performed to obtain a grayscale image of the infrared image. After the ordinary image processing step, the crack area and the road surface area (the area except the crack in the image) can be obtained, and then the RGB average values of the two areas are respectively calculated, and matched with the legend to obtain the temperature values of the two areas, thereby calculating the crack and Temperature difference data of the road surface. As shown in Figure 4 below:
  • the image needs to be processed to obtain the temperature data of the asphalt pavement crack and the road surface, and then further analysis.
  • the existing mature algorithm can be used to obtain the crack region and the road surface region, and then the RGB average value of the image in each region is obtained, and finally the average RGB value is matched with the colorbar, and sequentially from left to right. Stepping two pixel widths to ensure the accuracy of the temperature value, as shown in Figure 5 is the infrared image processing technology and the crack and road surface temperature acquisition technology route, Figure 6 is a schematic diagram of the image processing process.
  • the unprocessed crack image will contain noise information. If these noises are not removed, it will be difficult for subsequent image processing and result analysis. Therefore, image preprocessing is necessary. Preprocessing can remove some redundant information in the image and highlight the target we are interested in, so as to reduce the amount of image information and improve the image quality. First, the image needs to be noise-reduced.
  • the crack target to be identified in the road crack image has less information, and the image clarity and contrast are also reduced due to interference of many factors during the image acquisition and transmission. Therefore, after filtering and denoising the image, the graphics are further enhanced, so that the crack target we are interested in is more prominent, providing a basis for the latter segmentation recognition algorithm.
  • Image segmentation is the process of classifying pixels in an image into specific regions with unique properties and extracting the target's techniques and processes. Since people study images, they all have a goal, which is the area of interest in the image, and often these target areas have some specific properties.
  • the purpose of this step of the present invention is to find the crack area and the non-crack area of the road surface.
  • the average RGB value of each area can be matched with the colorbar to obtain the temperature value of the area.
  • the temperature data can then be processed.
  • the temperature difference is plotted on the ordinate and the light intensity is plotted on the abscissa.
  • the temperature difference is approximately the same as the atmospheric temperature, the temperature difference between the severe crack and the road surface is different.
  • the temperature difference is approximately the same as the atmospheric temperature
  • the temperature difference between the severe crack and the road surface is different.
  • the temperature of the road surface will increase as a whole, and the difference between the road surface temperature and the atmospheric temperature will be expanded.
  • the temperature is 35 ° C
  • the asphalt road temperature can reach 55 ° when the light intensity is 150,000 Lux. C ; however, the effect of illuminance on the asphalt pavement and the temperature of its surface is almost synchronous, so the illuminance has little effect on the temperature difference between the crack and the pavement.
  • the present invention only considers the relationship between the temperature difference between the crack and the road surface and the atmospheric temperature. At the same time, it can be seen from the above figure that for different degrees of cracks, the temperature difference is different at different temperatures, so we can reflect the degree of crack development by detecting the temperature difference between the crack and the road surface.
  • the traditional crack analysis technology based on image analysis only pays attention to the crack itself. It is necessary to detect the presence or absence of cracks, location and geometry, and the transition between the crack and the road surface is basically not considered.
  • the invention mainly needs to obtain the temperature difference between the road surface area and the crack area. It is necessary to pay attention to the two areas of the crack and the road surface. Therefore, the influence of the transition area can be considered. After the crack and the road surface area are obtained by image segmentation, the transition of the two area boundaries is required. The area is cut off because the temperature of the pavement to the crack temperature is gradual. The temperature of the transition zone between the crack and the pavement is between the crack center temperature and the pavement temperature. The transition zone is the interference zone for the pavement temperature, as shown in Figure 10. The following measures can be taken to exclude the interference zone.
  • the transition interference zone is small relative to the road surface area, so the road surface area can be cut at a fixed width when the exclusion is performed, where the width can be set to w width (0.5 cm ⁇ w ⁇ 2.5cm), in fact, it is also possible to adopt a method based on the maximum width ratio of the crack region, that is, after detecting the crack region, the crack region is doubled up and down, and the remaining region is defined as a pavement region without interference zone, assuming an image
  • the width of the divided road surface is 3 ⁇ 4, and finally the determined width is D, then the following formula (6) is used. This method can better eliminate the influence of the transition interference zone.
  • the ratio of the upper and lower parts of the crack width to eliminate the transition interference zone 0.1 ⁇ ⁇ 0.2.
  • how to delete it specifically can refer to the following: For the road surface area, it is equivalent to moving the image segmentation boundary along the segment boundary to the road surface area to obtain a new road surface boundary.
  • For the crack area The image segmentation boundary is moved along the segmentation boundary radially toward the crack region to obtain a new crack region boundary.
  • Asphalt pavement refers to various types of pavements paved with road asphalt materials in mineral materials. Asphalt binder improves the ability of paving granules to resist road and natural factors against road damage, making the road surface smooth, dust-free, impervious, and durable.
  • the road structure is a band structure built in nature, environmental factors and loads. The effect is the main reason for the damage of the pavement structure.
  • the physical characteristics of the pavement area are relatively consistent. Therefore, in the infrared image, the temperature value of the pavement area is basically the same, so it is convenient to handle, and no separate consideration is needed, that is, the pavement is obtained in the image segmentation. After the area is returned to the infrared image to calculate the RGB average of the entire area.
  • the crack area is usually elongated, such as a two-meter-long crack, which may be only a few millimeters wide.
  • the cracks are half-meter long and the development is more serious.
  • the average RGB value it may need to be repaired.
  • the half-meter-long area is more developed, and the average RGB value is calculated.
  • the development degree index obtained by the temperature difference will be lower than the previous one, and even shows that it is not repaired. It is obviously unscientific. Therefore, it is necessary to calculate the RGB value for the fracture area and give the area with more serious development.
  • the higher the weight, the calculated temperature difference between a crack area and the road surface will be an array, as shown in Figure 11, different sections of the crack can get different temperature difference.
  • the effective crack region is divided into the p segments by the straight line along the y-axis direction, and each length is arbitrary; for the longitudinal crack, the effective crack region is divided into the p segments by the straight line along the X-axis direction of the image, each length Any; for other types of cracks, not segmented or divided into segments according to the geometric center of the crack, each segment corresponds to the center angle.
  • the crack region divided by the image can be divided into "( « ⁇ 2) segments for consideration.
  • Each segment can be processed according to the technical route described above, that is, image graying and noise reduction are first performed.
  • the image is enhanced and then the image is segmented, and the crack region is obtained and then divided into “segments for subsequent processing, including calculating the average RGB value for each segment, and then matching the average RGB value with the color value in the legend to determine the segment.
  • ( ⁇ 1 ⁇ 2 ⁇ ⁇ 7 ) (9)
  • ⁇ 7 is the temperature difference between the / segment and the road surface after the fracture zone is divided into sections.
  • a weighted average calculation can be performed according to the following formula (10) to obtain a final measured temperature difference ⁇ 7 1 between the effective road surface region and the effective crack region.
  • the invention divides the degree of crack development into three levels of 1, 2 and 3 by machine learning, SVM support vector machine, as a classification method, through the kernel function
  • the linear indivisible samples in the low-dimensional space are mapped to the linearly separable sample space in the high-dimensional space, and the inner product is calculated by the kernel function to obtain a linear classifier.
  • kernel functions such as linear kernel functions, polynomial kernel functions, radial basis kernel functions, Sigmoid kernel functions, and composite kernel functions.
  • the linear kernel function is first used to classify the 1 and 2 grades to obtain the classification function / 12 , and then classify the 2 and 3 grades according to the data to obtain the classification function / 23 .
  • There is a certain error in the machine learning classification and the classification is performed.
  • some of the cracks that belong to the less developed degree are assigned to the category of more serious development, and the cracks with the heavier development degree are assigned to the lighter category.
  • there is a small number of fracture development index in the calculation of the average development index of the weights because there is a certain error in the classification of these.
  • these developmental indexes with decimals do not necessarily make the fracture development index more Precise, but can reflect the relative size, 2.
  • the development index of the 6 is divided into cracks with a developmental index of 3, and 2. 4 is divided into cracks with a developmental index of 2, which is the previous classification.
  • the method but the 2.6 developmental degree index and the 2.4 developmental degree index calculated by the weighting are actually not much different, which has a considerable influence on the decision-making of the manager. Therefore, in the calculation of the crack development degree index, the invention can take a decimal point to indicate the calculated relative relationship.
  • the developmental index of 2.4 is not necessarily more serious than the developmental index of 2.3, because as mentioned above, there is an error in the classification itself. 2 This integer part may already be inaccurate. .
  • the invention determines an error term according to the variance of the weighted average calculation method in the calculation process, that is, the development degree index can take one decimal place and is expressed as: m ⁇ .
  • the data is linearly classified using a support vector machine. Taking the atmospheric temperature as the abscissa and the temperature difference between the crack and the road surface as the ordinate, there are three levels of 1, 2, and 3. The larger the number, the more serious the development degree, and the following two classification functions can be obtained.
  • ⁇ 12 ⁇ 12 ⁇ +6 12 ( 15)
  • ( °C ) is the atmospheric temperature
  • ⁇ CO is the temperature difference between the asphalt pavement and the crack
  • 12 is the linear classification function coefficient
  • the value range is 0.0075 ⁇ 0.0100
  • 2 linear classification function constant term the value range is 0.4 ⁇ 0.65.
  • the degree index, which is greater than 3, is 3.
  • the result calculated by the formula can take one decimal place.
  • ⁇ ⁇ ' is the result of multiple tests on the same crack, and then the standard deviation is obtained from the above formula. According to this, the development index of the crack can be written to one decimal place, and the error is defined by a fixed error term. The fixed error term is obtained experimentally, and the new developmental degree index is written as follows (17):
  • Image acquisition can directly perform two-way video capture, and then the image processing technology directly processes the video stream in real time to obtain disease information of the road, and refers to space-time coordinates.
  • the above image acquisition methods have higher performance requirements for the processor, and the image processing algorithm is more complicated, but the requirements for the device are less, and it is not necessary to accurately measure the speed of the mobile platform.
  • Another image acquisition method needs to be coordinated by the photoelectric encoder and the synchronous controller.
  • the photoelectric encoder accurately acquires the speed of the mobile platform.
  • the synchronous controller determines the image shooting at a certain distance from the platform speed, and further triggers the two cameras simultaneously. Image acquisition is performed, and then the image data is transmitted to the processing end, and the processing end directly processes each picture to obtain disease data and store the diseased picture.
  • This type of acquisition reduces image algorithms and processor requirements by reducing the complexity of the algorithm by adding hardware.
  • Photoelectric encoder which is a sensor that converts the mechanical geometric displacement on the output shaft into a pulse or digital quantity by photoelectric conversion. It is the most widely used sensor.
  • a general photoelectric encoder is mainly composed of a grating disk and a photoelectric detecting device. In the servo system, since the photoelectric code disk is coaxial with the motor, when the motor rotates, the grating disk rotates at the same speed as the motor.
  • the detecting device composed of electronic components such as light-emitting diodes detects and outputs a number of pulse signals. The current motor speed can be reflected by calculating the number of photoelectric encoder output pulses per second.
  • the encoder can also provide two channels of optical code output with a phase difference of 90°, and determine the steering of the motor according to the state change of the two-channel optical code.
  • the encoder can be divided into optical, magnetic, inductive and capacitive. According to its scale method and signal output form, it can be divided into increments. 3 types of formula, absolute type and hybrid type.
  • Synchronous controller refers to the coordination of the position, rotation speed and torque between the master and the slave according to a certain ratio.
  • the tension controller is also a synchronous control device. This type of synchronization is realized by the synchronization of the same speed and torque.
  • the other type is the space positioning controller.
  • Position synchronization generally used in the linkage between the axes of robots, CNC machine tools, flying shears, etc., is a kind of position tracking between axes.
  • the synchronous controller has embedded setting parameters, and also has direct programmable class. With the development of technology, the application of programmable class slowly surpasses the former, and the development direction of the representative synchronous technology can be through the field bus. Communication technology and other devices are connected and operated.
  • the temperature information of pavement and crack is used to detect the degree of crack development.
  • the temperature of the pavement changes greatly and is easily interfered by environmental factors.
  • the requirements for the experimental environment are higher than the crack recognition based on ordinary images.
  • We determine the environmental requirements of the experiment based on the law of changes in the temperature of the road surface. The environmental requirements ensure the reliability of the final model.
  • the maximum temperature of the road surface appears almost simultaneously with the daily maximum temperature.
  • the temperature at each depth of the road surface is always approximately equal.
  • the lowest temperature of the road surface also appears almost simultaneously with the daily minimum temperature.
  • the time at which the daily minimum temperature occurs at different depths gradually lags behind. That is, although the solar radiation has a similar law to the temperature of the road surface, the influence on the road surface temperature is hysteretic and cumulative.
  • FIG. 1 Modeling system sampling data, including infrared image (101), atmospheric temperature (102), crack development degree (103), measured temperature difference data (104); Figure 2 Schematic diagram of the fixing device of the collecting device;
  • FIG. 3 Schematic diagram of the acquisition end and the processing layer
  • FIG. 4 Schematic diagram of the infrared image temperature difference
  • Figure 7 shows the relationship between illumination and temperature difference
  • Figure 8 shows the relationship between atmospheric temperature and temperature difference
  • Figure 9 Support vector machine linear classification map
  • FIG. 11 Schematic diagram of different temperature differences obtained for different crack sections
  • FIG sample data application stage 12 including an infrared image (201), air temperature (202), a normal image (203), the vibration information (204), space-time coordinates of the TS ⁇ TS 2 and TS 3.
  • the predictive relationship model needs to be used under certain environmental conditions to ensure accuracy. First, it is necessary to ensure that the environmental conditions are stable during data collection.
  • the experimental environment can be met: Sunny daytime, effective data collection time is 8:00 am to 4:00 pm, and the road surface is completely Dry and clean.
  • the ordinary camera (with night vision function) and the infrared camera (temperature measurement type) are fixed on the mobile platform, and the road inspection is carried out in the collection environment, first select a specific lane, and the initial lane information is recorded, and the inspection speed is the highest. Up to 80km/h, during the inspection, the two cameras perform real-time image acquisition, the frequency is greater than or equal to 60Hz, the vibration sensor acquisition frequency is at least 100Hz, the real-time high-frequency sampling is performed along the line acquisition device, and the positioning device and system time are used for real-time spatial and temporal coordinate acquisition. .
  • the use scenario is an asphalt pavement for each grade of road. This phase belongs to the application phase, and the specific sampling data is shown in Figure 12.
  • each road is treated with a road disease identification algorithm, and the images are classified and stored in a disease-free manner, and the types of diseases are further identified for the diseased images, and the crack-type diseases are further based on infrared images. Identify the degree of crack development, and then form the log information of the disease to match the positioning information. The same crack may exist in multiple images.
  • each image will receive disease information. After matching with the positioning information, the disease can be determined. The uniqueness, that is, only one of the same disease types at the same location coordinates. Finally, the disease information is uploaded to the central server for display and application.
  • the vibration signal is processed, and the flatness information of the road segment is calculated in real time by the high-frequency vibration signal, and matched with the space-time coordinates to form a complete flatness detection information, which is then uploaded to the server database for display and application.
  • the first layer is the display release of the job layer.
  • the data release layer is configured on the vehicle side.
  • the display terminal is directly connected to the processor.
  • the data is derived from the real-time processing end to transmit information, mainly to visualize the data stream. Display and display the working status display of the working condition, no interactive interface.
  • the working conditions are divided into normal, abnormal and error.
  • the display of the data flow assists the relevant personnel to eliminate the abnormal working conditions.
  • the wrong working condition is a device problem and needs to be checked for downtime.
  • the abnormal working conditions are mainly the abnormal collection environment conditions such as unsatisfied climatic conditions or too fast moving platform.
  • the second layer is the release of the application layer, mainly the visual web form, which is divided into a log interface and a visual map interface.
  • the log page can realize the operation of querying, exporting and importing data.
  • the visual interface uses the position and information of the disease on the two-dimensional map. , you can switch to the flatness display mode at the same time.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • General Engineering & Computer Science (AREA)
  • Architecture (AREA)
  • Civil Engineering (AREA)
  • Structural Engineering (AREA)
  • Mechanical Engineering (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Medicinal Chemistry (AREA)
  • Food Science & Technology (AREA)
  • Acoustics & Sound (AREA)
  • Image Processing (AREA)
  • Investigating Or Analyzing Materials Using Thermal Means (AREA)
  • Radiation Pyrometers (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

一种基于双目图像分析的沥青路面病害检测系统,包括裂缝发育程度检测模型建立、信息采集、信息分析、信息传输和信息发布五个子系统,具体通过移动平台、处理端、测温型红外热像仪、普通图像采集器、振动传感器、图像处理技术、边缘云计算技术、嵌入式系统、系统集成技术、实时定位技术来进行路面病害信息检测。

Description

一种基于双目图像分析的沥青路面病害检测系统
技术领域 本发明属于智能交通系统与信息技术领域, 涉及运用裂缝发育程度检测模型建立、信息采 集、 信息分析、 信息传输和信息发布五个子系统, 具体包括移动平台、 处理端、 测温型红外热 像仪、 普通图像采集器、 振动传感器、 图像处理技术、 边缘云计算技术、 嵌入式系统、 系统集 成技术、 实时定位技术来进行路面裂缝病害信息检测, 信息包括裂缝的数量、 类型、 位置和发 育程度。通过引入红外热图像可以获取沥青路面裂缝的灰度信息和温度信息, 灰度信息可以实 现常规基于图像的裂缝识别的目标, 而利用温度信息可以进行裂缝发育程度的检测, 利用上述 技术实现裂缝检测系统在移动平台上的搭建, 实现沥青路面裂缝病害的快速准确检测。 背景技术 随着道路交通的发展,道路路面的养护和管理问题愈发凸显, 其中道路路面的破损检测就 成为相关道路养护部门的工作重点之一。目前我国的路面管理系统得到的路面破损数据仍以采 用人工检测为主, 效率低且受主观因素干扰。 随着近年来数字图像处理技术的飞速发展, 国内 外许多研究人员都对图像识别路面裂缝进行了广泛研究, 取得了很好的成果。然而, 现有路面 病害检测车采用多维全面高精尖的技术,检测车昂贵且使用成本高, 难以实现各等级道路的全 覆盖高频巡检, 造成智慧养护的决策支持数据不足, 因此, 道路的智慧管养亟待可以实现各等 级公路全覆盖的高频巡检设备, 以便采集充分的损害数据给予决策支持, 实现整个路网的智慧 管养。
( 1) 南京理工大学 N-1路面检测车
在 2002年南京理工大学赵春霞、 唐振民等成功研制了 N-1型路面检测车。 该检测车可在 70km/h行驶状态下, 完成表面裂缝图像、 平整度、 车辙等数据的采集, 再离线处理数据, 生 成报表。使用车载全球定位系统和车载前视彩色摄像机。该车的诞生推动了中国路面自动检测 车的发展进程。
( 2) 公路科学研究院 GCS路面检测车
2006年,北京公路科学研究院下属子公司: 中公高科养护科技股份有限公司,研发了具有 完全自主知识产权的路面检测车 CiCS。该系统采用线阵相机采集路面图像,并结合结构光照明, 能够达到横向 3.6米的检测宽度, 最高检测精度达 lmm, 最高时速达 100km/h。该公司研发的 路面破损处理软件 CiAS (Cracking image Analysis System), 能够自动处理包括裂缝、 车辙以及平 整度在内的路面关键性指标。 ( 3 ) 武大卓越 ZOYON-RTM路面检测车
2007年, 武大卓越与湖北合力联合研发了 ZOYON-RTM型智能道路检测车,。 该车使用高 精度线阵相机对路面图像进行采集, 并采用了激光照明技术。在编码器的同步触发信号控制下 实现路面图像的不间断采集。 该系统的裂缝分辨率最高为 lmm, 最大检测宽度达到了 4m, 最 高时速 120千米每小时。 现有技术
专利文件 CN200910222882.5 ,公布了一种基于图像的路面裂缝检测系统,该系统包括:控 制模块, 根据接收到的 GPS定位信息生成触发信号, 输出触发信号至图像至采集模块; 图像采 集模块, 根据接收到的触发信号对路面及路标进行图像采集获得路面图像及路标图像,将路面 图像和路标图像输出值数据处理模块; 数据处理模块, 将第一数字图像拼接获得路面的连续的 数字图像,利用图像识别方法对路面的连续的数字图像中的裂缝进行识别获得包含裂缝的第一 数字图像、 GPS定位信息及第二数字图像, 根据 GPS定位信息及第二数字图像确定路面裂缝的 位置信息并输出。发明还提供理论一种基于图像的路面裂缝检测方法。采用发明公开的系统和 方法, 能够题号检测的准确度和检测效率。 现有技术 2
专利文件 CN201310252824.3 ,公布了一种路面裂缝检测系统,包括有信息采集系统、数据 存储显示系统、 电源管理系统、 数据处理系统、 GPS定位系统。 通过信息采集系统视频记录路 面破损信息, 结合 GPS定位系统的 GPS和公路里程桩号对病害位置进行定位, 利用数据存储 显示系统的数字图像处理技术对病害进行识别, 既可得到客观稳定的检测结果,又能快速准确 地对病害进行定位、 测量。本发明针对路面裂缝检测工作, 实现完成了相关信息实时监测、 数 据存储回放、 GPS信息与里程桩号实时关联、裂缝状态信息度量等需求,设计了车载检测系统, 满足了实际路面裂缝检测的应用要求。 现有技术 3
专利文件 CN201410269998.5,发明公开了一种基于深度和灰度图像的路面裂缝检测装置, 包括: 载体平台 (1) ; 位于载体平台上摄像装置 (6)、 线激光器 (7)、 计算装置 (8), 其中, 载体平台(1)用来在裂缝检测的过程中,沿道路方向移动,线激光器(7)用来在载体平台(1) 移动的同时, 垂直路面照射激光, 摄像装置(6) 用来在载体平台 (1) 移动的同时, 沿斜向角 度, 不远拍摄所述直线激光经过路面反射的激光线, 每次拍摄对一个道路断面进行成像, 生成 多个道路断面的激光线的图像数据, 计算装置 (8) 用来从每个道路断面的激光线的图像数据 生成该道路断面的深度数据、 以及灰度数据, 并将每个道路断面的深度数据和灰度数据拼接形 成一段道路的图像数据, 以进行裂缝识别。 为了更好的理解发明内容, 现在附上以下名词解释:
名词解释:
裂缝发育程度:发明中所提及的发育程度表征沥青路面裂缝对路面的已有损害程度以及近 期发生损害的严重程度, 它涵盖了传统的裂缝严重程度分级, 与裂缝的宽度、深度和裂缝的湿 度等等可能会使裂缝损害严重程度加剧的因素有关, 囊括了这些因素,表征了裂缝从开始出现 到现在的发展水平。
分类函数:通过支持向量机将温差数据按照裂缝发育程度进行线性分类,发育程度分为 1、 2、 3三个等级, 3最严重, 那么 1、 2之间和 2、 3之间就会有一直线作为分界线, 此直线表达 式即为分类函数。
裂缝区: 沥青路面某区域, 该区域不仅包括裂缝区域本身, 还包括其周围一定范围内的路 面区域, 包括路面区域范围满足图像处理和裂缝识别的要求。
参照温差数据:将实测大气温度带入裂缝发育程度检测模型两个分类函数后得到的温差数 据。
实测温差数据:对采集到的裂缝区红外图像进行图像处理后得到的该裂缝区内裂缝区域和 路面区域的温差数据。
发育程度等级: 1、 2或 3中的一个数, 其大小反映了裂缝的发育程度, 数字越大, 裂缝发 育程度越严重。
处理端: 负责接受采集到的原始数据, 并进行存储和实时处理, 最后再将处理得到的数据 无线传输到远程服务器。
支持向量机: 支持向量机 (SVM ) 是 90年代中期发展起来的基于统计学习理论的一种机 器学习方法,通过寻求结构化风险最小来提高学习机泛化能力, 实现经验风险和置信范围的最 小化, 从而达到在统计样本量较少的情况下, 亦能获得良好统计规律的目的。 在机器学习中, 支持向量机(SVM, 还支持矢量网络) 是与相关的学习算法有关的监督学习模型, 可以分析数 据, 识别模式, 用于分类和回归分析。给定一组训练样本, 每个标记为属于两类, 一个 SVM训 练算法建立了一个模型, 分配新的实例为一类或其他类, 使其成为非概率二元线性分类。通俗 来讲, 它是一种二类分类模型, 其基本模型定义为特征空间上的间隔最大的线性分类器, 即支 持向量机的学习策略便是间隔最大化, 最终可转化为一个凸二次规划问题的求解。
病害识别模型算法: 根据沥青路面灰度图识别裂缝有无、 类型的成熟算法, 普通图像和红 外图像均可以得到灰度图。 发明内容 本发明提出基于红外双目图像分析和功率谱密度的路面病害检测系统,系统包括发育程度 检测模型建立、 信息采集、 信息分析、 信息传输和信息发布五个子系统。
普通图像只能获得沥青路面和裂缝的灰度信息,普通图像难以较好识别沥青铺面裂缝的原 因是沥青路面具有颗粒性, 灰度的杂乱往往会影响裂缝区域的识别。我们通常通过识别裂缝的 宽度来检测裂缝的发育程度, 而裂缝的宽度是通过识别裂缝区域短边的像素点个数, 并结合摄 像机机位拍摄裂缝时的高度来确定。然而, 通过裂缝像素点宽度的识别来得到裂缝宽度进而确 定裂缝发育程度的检测方法很不可靠,因为几毫米的的裂缝宽度在图像中可能就只有几个像素 点, 在经过降噪等图像处理后结果更不稳定, 所以, 只由普通图像获取的裂缝的灰度信息进行 裂缝发育程度的识别是很困难的。
由红外热像仪获取的红外热图像,不仅可以获取裂缝和铺面的灰度信息还可以得到他们的 温度信息
由道路表面温度滞后性和积累性的特点, 路表和路面以下的温度会不同, 因此路面以下的 材料会通过裂缝与路面和空气发生热交换, 而路面和裂缝也会有温度差。沥青路面裂缝宽度和 深度越大, 路面下和路表的材料空气热交换救会越剧烈。温差就会越大。裂缝的宽度深度就越 大, 将来的水损害也会更容易发生, 而裂缝也通过发育程度指数反映发育程度。所以, 裂缝越 严重, 路表与裂缝的温差就会越大, 而我们可以利用热像仪来检测温差, 进而检测裂缝以及其 发育程度。 检测方法采用基于红外图像分析的裂缝发育程度检测模型。
车辆行驶过程中的加速度变化主要由路面高程的变化产生,因此将路面高程变化看作系统 激励, 车身的加速度看作系统响应, 利用功率谱密度和线性时不变系统理论, 能够推导出国际 平整度指数(IRI )与簧载加速度功率谱密度的转换公式, 通过测量车内不同位点的加速度值能 够提高计算精度, 而车速修正系数能够修正车速变化对计算结果的影响。在桥头跳车处, 基于 车内不同位点测得的加速度值, 结合人体振动感受函数和烦恼率, 计算得出桥头跳车对人体舒 适性的影响, 并利用冲击系数和系统响应的方法预估桥头当量冲击力大小。利用系统响应的方 法, 通过测量车内的 Z轴加速度变化, 间接预估路面平整度的检测模型。 该方法具有便捷性、 低耗性以及经济合理性等优点, 适合大范围的路面平整度测量。 另外, 利用小波理论、 卡尔曼 滤波、 线性时不变系统等成熟的数学方法, 以及实验心理学中烦恼率等概念, 可以对桥头跳车 现象进行多方面的评估。
考虑到本实验需要建立 Z轴加速度与 IRI之间的相关关系, 因此实验中需要通过三轴重力 加速度传感器进行实地的测量以采集数据,
因此, 为了定量探讨速度对模型结果的影响, 我们选择不同速度在同一试验路行驶, 用来 观察测量结果差异, 实地试验后发现速度和功率谱密度之间有着很好的二次相关关系,表达如 下式 (1) :
PSD = 0.0263v2 + 0.6027v ( 1) 且 R2 = 0.9991, 表明拟合优度极高, 则将该表达式带入上文推导的 IRI预估公式中, 可得 到最终的 IRI表达式 (2 ) :
IRI = 0.782axfe i + l .300 xri ht - 3.442 .
V0.0003865V2 + 0.0009125V
采用基于红外热像图的路面裂缝发育程度检测方法进行检测系统的构建,实际的检测系统 需要利用红外双目摄像机进行路面图像的采集。双目摄像机包括一个普通相机和一个红外热像 仪, 同步进行沥青路面的图像采集, 并进行处理进行裂缝的有无、 位置、 大小和发育程度的识 别, 两种图像数据融合, 可以提高发明提出的检测方法的鲁棒性。
裂缝检测系统采用模块化设计, 可以很好的与其它路面病害检测模块进行结合。裂缝检测 模块包括摄像机固定装置、 双目摄像机、 数据传输线、 车载终端、 GPS 接受装置、 惯性导航。 摄像机固定装置可以采用定制铁架,保证可以在工程车上可靠固定双目摄像机; 可以采用车载 云台, 固定摄像机的同时可以进行双目摄像机视角的控制, 云台至少拥有两个自由度, 即可以 分别在水平方向和竖直方向旋转, 更多自由度将带来云台的更高操作性。红外摄像头的最低配 置为: 分辨率为 320 X 240, 带可更换的镜头, 免维护的非制冷微测辐射热计, 显微技术和 特写镜头测量功能,数据传输速度高达 60 Hz。普通图像摄像头支持 1080P高清图像实时输出。 数据传输线包括至少两根,分别支持高清普通图像和红外图像的高频传输,支持至少 100Hz的 1080p视频传输。 车载终端包括两种方案: 一种是前端处理, 采用嵌入式 PC进行实时视频流 的处理, 通过 3g\4g网络进行处理后的数据传输, 嵌入式 PC可以兼容普通图像和红外图像的 数据接口, 处理器支持视频流的实时处理; 另外一种是后端处理, 车载终端作为数据采集的前 端, 只负责数据采集和存储, 配置和开发要求更低, 需要有接口供采集后第三方设备接入进行 数据处理。此种模块通过视频图像拼接技术, 还原整个采集过程的路面纵向图像, 然后再沿长 度方向进行切割后分别识别,保证沥青路面裂缝检测的准确性和稳定性。此系统思路图像处理 算法要求高, 设备相对要求更少。 GPS 接受装置和惯性导航, 两者共同为采集设备进行定位, 确保精度和实时性。 定位的精度需要在 10m内。
另外一种裂缝检测模块包括摄像机固定装置、 双目摄像机、 数据传输线、 车载终端、 GPS 接受装置、 惯性导航、 光电编码器和同步控制器 (参照 CN104749187)。 光电编码器安装在车 载移动平台的车轮中心轴上, 用以测量车载移动平台的运行速度和距离; 所述 GPS接收机, 安 装在车载移动平台上, 用于所述车载移动平台的高精度定位及授时; 所述惯性导航, 安装在车 载移动那平台上, 用于在隧道内 GPS接受机接受不到 G PS信号的情况下, 测量车载移动平台 的位置、姿态数据,实现在隧道内部高精度的位置推算;同步控制器,安装在车载移动平台上, 用于同步普通摄像机和红外摄像机的图像采集时间,保证两者具有统一的时间和空间基准。此 种方法通过光电编码器精确测量车速, 同步控制器根据车速和双目摄像机的视野大小, 自动控 制双目图像采集时间,确保相邻两张有效剪辑的路面图像具有很好的连续性, 可以完整的覆盖 所采集的车道, 并且相互不重叠。 图像传输到车载终端后执行前端处理或者存储功能, 此系统 思路对于设备要求较高, 图像处理算法要求较低。
裂缝发育程度检测模型建立子系统:
采集沥青路面至少 10个样本裂缝区的红外图像, 同时记录样本大气温度及所述样本裂缝 区内的裂缝发育程度, 将图像处理后得到采样样本的样本温差数据。 以大气温度为横坐标, 相 应红外图像中的裂缝区与路面区的样本温差数据为纵坐标绘制坐标点。具体在模型建立阶段需 要的采样数据如图 1所示。
采用支持向量机对坐标点进行线性分类。 以大气温度为横坐标、裂缝与路面温度差为纵坐 标绘制点, 共有 1、 2、 3三个等级, 数字越大发育程度越严重, 如图 8所示为分类函数图, 可 以得到如下两个分类函数, 如式 (4) ( 5 ) 所示。
3、 2分类函数: ΔΓ23 = 3Γ+ 3 (4) 其中 ( °C ) 为大气温度, Γ CO 为沥青路面与裂缝的温度差; α23为线性分类函数 系数, 取值范围为 0.02~0.03, 3线性分类函数常数项, 取值范围为 0.60~0.85.
2、 1分类函数: ΔΓ12 = α12Γ+612 ( 5 ) 其中 ( °C ) 为大气温度, Γ CO 为沥青路面与裂缝的温度差; 12为线性分类函数 系数, 取值范围为 0.0075~0.0100, 2线性分类函数常数项, 取值范围为 0.4~0.65. 检测结果根据以下判断 首先根据大气温度计算出 ΔΓ12和 ΔΓ23, 然后将测量出的 Δ 与 ΔΓ12、 ΔΓ23进行对比,
AT≤ ATU则发育程度为 1; ΔΓ12 < ΔΓ < ΔΓ23则发育程度为 2, AT≥ AT23的发育程度为 3. 现在在其它沥青道路上进行数据采集, 验证上述检测模型的精度。 由上文分析, 在路面 完全干燥的情况下, 沥青路面与裂缝的温差主要与温度有关系, 并且裂缝与路面的温度差和 裂缝的发育程度是相关的, 因此可以由红外热像仪检测裂缝与路面的温度差, 再利用上文中 分类函数 /^π /2, 进行裂缝发育程度的检测。 首先确定检测环境满足要求, 即前文所述实验环境: 晴朗的白天, 上午八点到下午四点 之间, 路面干净并完全干燥。 然后利用红外热像仪进行数据采集, 并记录下采集每张图片时 的气温, 处理后得到裂缝与路面的温差, 根据温度计算发育程度分级阈值 ΔΓ12 和 ΔΓ23。 表 2 模型验证结果 裂缝等级 实际数量 理论数量 相对误差
轻 28 23 17.8%
中 17 15 11.8%
轻 12 10 16.7% 如表 2所示, 为检测模型验证的结果, 实际检测的相对误差均小于 20%, 模型检测的平均 误差为 15.4%.结果表明预测模型的精度较好,在目前大多裂缝检测都只关注数量的背景下, 可 以根据检测模型结果赋予不同的权值给不同发育程度的裂缝, 为路面养护提供更精准的参考, 提高社会效益。
信息采集子系统
数据采集层包括车载定位设备、 主动红外图像传感器、测温型红外热像仪、三轴加速度传 感器。
车载定位设备: 车载移动测量系统由于其方便快捷、 测量效率高等的特点, 使其在城市规 划、 道路巡检、 数字城市等诸多工程中得到应用, 成为了目前测绘的一个重要方向。 为了实现 高精度的车载移动测量,需要对全球卫星导航定位系统 (Global Navigation Sat-elite System, GNSS) 定位结果提出更高的要求。由移动测量系统的特点决定了其车载 GNSS定位需要高精度、高频率 、快速的实现动态定位。目前,随着我国北斗卫星导航系统 (Bei Dou Navigation Satellite System, BDS)的建成运行、俄罗斯 GLONASS系统的大力复兴与美国 GPS系统的现代化,空中将会有更多 的高质量导航卫星。多系统组合观测可大大增加观测卫星数量,明显改善卫星的几何分布。本发 明采用最新的基于北斗或 GPS的定位技术,实现车辆位置信息的精确实时刷新,设备可以在一 定频率范围内持续向处理终端有线传输位置信息, 包括时间和经纬度。
红外摄像机,尽量避免直射光源, 因为红外灯电源控制部分是根据安装在红外灯板上的光 敏电阻来控制红外灯的工作电源开启与否的。 红外摄像机视场内应尽量避免有全黑色物体、 空旷处、 水等吸收红外光线的物体, CCD摄 像机配套的红外灯是靠发射红外光在物体上反射到 CCD摄像机镜头上成图像的, 如果红外光 被吸收或减弱, 将会大大地削弱红外灯的有效照射效果。
主动红外图像传感器设备实现检测路面的灰度图像采集, 图像质量至少达到 720p, 支持 数字图像有线传输至数据处理端。
测温型红外热像仪: 直到 20世纪六十年代, 热成像技术才被用于非军事应用领域。 虽然 早期的热成像系统很笨重、数据采集速度缓慢而且分辨率不佳, 但它们还是被用于工业应用领 域,例如检查大型输配电系统。 20世纪七十年代, 军事应用领域的持续发展造就了第一个便携 式系统。 该系统可用于建筑诊断和材料无损测试等应用领域。
20世纪七十年代的热成像系统结实耐用而且非常可靠, 但与现代热像仪相比, 它们的图 像质量不佳。 到 20世纪八十年代初期, 热成像技术已广泛应用于医疗、 主流行业以及建筑检 查领域。经过校准后, 热成像系统可以制作完全的辐射图像, 这样便可测量该图像中任意位置 的辐射温度。 辐射图像是指包含图像内各点处的温度测量计算值的热图像。
安全可靠的热像仪冷却器经过改进,取代了沿用已久的用于冷却热像仪的压缩气或液化气。 此外, 人们还开发并大量生产了成本较低、 基于管道的热电光导摄像管(PEV)热成像系统。 虽 然不能进行辐射测量, 但 PEV热成像系统轻巧灵便、 携带方便, 而且无需冷却便可操作。
20世纪八十年代后期, 一种称为焦平面阵列(FPA)的新设备从军事应用领域转移至商业市 场。焦平面阵列(FPA)是一种图像传感设备, 由位于镜头焦平面处的红外传感探测器的阵列(通 常为矩形) 组成。
采用的测温热像仪参数可以参考以下但是不限于此:
空间分辨率: 0.68mrad;
像频: 50Hz(100/200Hz带窗口);
焦平面阵列 (FPA) /波长范围: 非制冷微量热计 /7.5~14um ; 红外分辨率: 640 X 480像素;
对象温度范围: -20°C~120°C ;
支持 USB和以太网图像流输出。 测温型红外热像仪需要实现检测路面的红外热图像采集,可以得到裂缝与路面的温度信息。 三轴加速度传感器: 三轴加速度传感器大多采用压阻式、 压电式和电容式工作原理, 产生 的加速度正比于电阻、 电压和电容的变化, 通过相应的放大和滤波电路进行采集。这个和普通 的加速度传感器是基于同样的原理,所以在一定的技术上三个单轴就可以变成一个三轴。对于 多数的传感器应用来看, 两轴的加速度传感器已经能满足多数应用。但是有些方面的应用还是 集中在三轴加速度传感器中例如在数采设备,贵重资产监测,碰撞监测,测量建筑物振动,风机, 风力涡轮机和其他敏感的大型结构振动。
发明采用集成设计加速度传感器,传感器模块具有无限传输模块和锂电池, 可以实现车辆 后轴振动信号采集, 频率最高可达 200Hz.
以上设备中,加速度传感器为平整度检测数据检测模块的采集设备、两种相机为裂缝等路 面病害检测的采集设备, 定位设备是为实现自动检测处理的辅助设备。各模块的采集设备和辅 助设备可以参照以下图 2进行设计,可以固定在各类移动平台上。车载固定架包括两个摄像机 平台和三条固定架,每支固定架上有 2~4个磁铁吸附装置, 可以使用三条支架上的磁铁吸附装 置进行两个摄像机平台的固定。 同时具有两个螺丝孔, 必要可以使用螺丝将采集系统固定在车 辆设备上。 支架需要保证摄像机的稳定性, 以及在车辆快速移动中的可靠性。
数据 H
数据传输层包括有线传输和无线传输两种方式。
有线传输包括两路视频和定位信息, 普通图像采用高清视频传输线, 红外图像可以采用高 清视频传输线或者网线进行传输。 振动信号的传输采用无线传输方式, 参考 zigbee或 WIFI方 式, 发明采用 zigbee传输方式, wifi作为技术储备。
简单的说, ZigBee是一种高可靠的无线数传网络, 类似于 CDMA和 GSM网络。 ZigBee数 传模块类似于移动网络基站。通讯距离从标准的 75m到几百米、几公里,并且支持无限扩展。 ZigBee是一个由可多到 65000个无线数传模块组成的一个无线数传网络平台,在整个网络范围 内,每一个 ZigBee网络数传模块之间可以相互通信,每个网络节点间的距离可以从标准的 75m 无限扩展。
与移动通信的 CDMA网或 GSM网不同的是, ZigBee网络主要是为工业现场自动化控制数 据传输而建立, 因而, 它必须具有简单, 使用方便, 工作可靠, 价格低的特点。 而移动通信网 主要是为语音通信而建立, 每个基站价值一般都在百万元人民币以上, 而每个 ZigBee "基站" 却不到 1000元人民币。 每个 ZigBee网络节点不仅本身可以作为监控对象, 例如其所连接的传 感器直接进行数据采集和监控, 还可以自动中转别的网络节点传过来的数据资料。 除此之外, 每一个 ZigBee网络节点 (FFD)还可在自己信号覆盖的范围内, 和多个不承担网络信息中转任务 的孤立的子节点 (RFD)无线连接。
数据传输以可靠性、实时性和稳定性为原则进行设计, 保证数据处理层与数据采集层的高 效通信, 同时需要将控制端的指令信息发送至采集设备。 网络传输如图 3所示。 数据存储设备内置与处理端, 主要对系统处理日志、 原始病害图像数据、 原始振动数据以 及他们对应的时空坐标进行存储, 以便后期进行进一步处理和大数据积累使用。
以上为采集层到处理层的数据传输,还有从处理层到服务器数据库的传输, 实时传输只针 对于数据处理后的病害信息, 主要使用无线网络进行传输, 包括 3g/4g网络, 并在数据量允许 的条件下使用窄带物联网传输技术。 此层数据传输只从处理端到服务器中心单向传输。 信息分析子系统
数据处理层包括普通图像处理、红外图像处理和振动信号处理, 通过图像处理得到有病害 的图像帧与位置信息匹配, 最终处理得到道路路面病害类型和位置; 通过对振动信号处理, 再 与位置信息匹配, 得到道路的沿线平整度和异常跳车信息。
普通图像处理:首先需要对图像进行灰度化和降噪处理, 图像降噪适用于普通图像和红外 图像, 现实中的数字图像在数字化和传输过程中常受到成像设备与外部环境噪声干扰等影响, 称为含噪图像或噪声图像。减少数字图像中噪声的过程称为图像降噪,有时候又称为图像去噪。 噪声是图像干扰的重要原因。 一幅图像在实际应用中可能存在各种各样的噪声,这些噪声可能 在传输中产生,也可能在量化等处理中产生。
图像降噪的方法主要有以下几类: 均值滤波器、 自适应维纳滤波器、 中值滤波器、 形态学 噪声滤除器、 小波去噪等。
边缘检测是各种图像检测应用中非常常见的一项检测流程,因为感兴趣特征通常其图像局 部灰度变化显著, 与背景图像有较大的差异, 在这些变化强烈的地方存在边缘。 图像的边缘信 息可以分解为边缘方向和边缘幅度特性, 通常边缘的幅值在沿着边缘方向上其变化比较平缓, 而在垂直边缘的方向上呈现剧烈的梯度变化。依据边缘的这种特性, 学者们提出了许多基于一 阶或二阶的微分算子来实现边缘的检测。对图像采用离散化的梯度逼近函数就可以找到图像中 梯度变换较大的位置,然后再将图像中将这些极值点连接起来即可以构成图像边缘,但是在这 个过程中, 需要对边缘的真伪信息做进一步的甄选。
本发明采用 Canny算子进行边缘检测来识别裂缝病害, 此为裂缝检测的核心算法。 Canny 算子的检测效果是最为优秀的, 其主要特点在于噪声的控制非常好, 能够将伪裂缝去除, 图像 比较干净, 形态学处理处理有大大优势。 Canny 算子的关键参数为 Alpha 的取值和两个高低 阈值参数。再通过对梯度幅值进行非极大值抑制减少伪裂缝和双阈值的分割和连接,使得边缘 连续成一整体且清晰可辨。
在边缘信息提取出来的后一个步骤便是阈值分割, 如果没有进行恰当的阈值分割,边缘信 息将会非常杂乱, 伪裂缝占据了整个图像区域, 所以阈值分割的步骤至关重要。 阈值分割的基 本思想是基于原始图像中的目标与背景在灰度值上的差异特性来进行区分的。通过设定一个或 多个阈值, 将图像划分为目标(前景)和背景两个部分。 阈值分割主要分为全局和局部阈值法 两种, 全局阈值法, 包括双峰法、 大津法、 最小误差法和一维最大熵法等等, 本发明采用一维 最大熵法。
经过图像的边缘检测和阈值分割后,得到的图像会存在一部分多余的伪边缘。下一步处理 需要将这些无效的信息去除,裂缝边缘的毛剌得到去除以及将间断的裂缝边缘连接起来。本发 明采用数学形态学的方法来解决这些问题。数学形态学是一门建立在集合理论的学科, 是以图 像的形态为基础,对图像进行近似数学分析的工具。其核心思想是用具有一定形态的结构化元 素去对原始图像进行度量和处理。数学形态学可以简化图像数据, 可以应用于图像分割, 噪声 抑制, 图像恢复等领域。数学形态学的基本运算操作有 4个:腐蚀(Erosion ^膨胀(Dilati0n)、 开启 (Opening) 和闭合(Closing), 前二者运算操作的组合可以形成后二者操作。 在处理的过 程中,结构元素相当于一个"滤波窗口",使用这个结构化的元素,将其与待处理图像进行交、 并等集合运算。腐蚀运算能够有效地去除分割图像中的孤点, 并对裂缝有细化的效果, 为了近 似的保持原来裂缝尺寸, 最后对裂缝图像进行了膨胀运算。
通过前述的处理后,裂缝的特征已经从背景图像中被分离出来, 然而在这些被分割出来的 区域通常都包含非裂缝噪声和其他干扰因素, 比如油渍, 杂物。 因此为了正确的提取裂缝, 对 于裂缝的判定是至关重要的一步,本发明利用一种基于融合的裂缝形态特征的选择方法, 能够 达到 90%以上的正确率。
( 1) 最小外接椭圆的长轴与短轴之比, 定义如下式 (3) 所示参数:
Anisometry = Ra /Rb ( 3) 裂缝在小块图像内通常呈现一个比较狭长的区域, 用椭圆将每一个待判定的裂缝 Region 进行外接, 其中求得区域内每一个 Region的 Anisometry参数, 并记录最大 Anisometry, 作为 判定依据。
( 2) 裂缝区域面积 Area
计算每一个待判定的裂缝区域的像素面积, 将其保存至 Area数组中, 并提取面积最大的 Region的 Area数据。 由于裂缝通常占据了分块面积中的较大一部分, 其余的伪裂缝的面积通 常较小, 这样裂缝区域的面积也可以作为一个判断是否为裂缝的依据。
发明首先介绍了将路面图像分块检测的思想,然后是对路面裂缝图像进行边缘检测, 在发 明中, 我们采用一种改进的 Canny算子, 对路面的裂缝边缘进行检测, 实验结果显示检测效果 较为优秀。在阈值分割阶段, 发明考虑到路面复杂状况的适用性, 选择了一维最大熵法来对图 像进行分割, 并做了对比实验。然后着重讲解了怎样去除被分割出来的裂缝孤点, 发明采用了 数学形态学来对孤点伪裂缝进行了提出, 取得了良好的效果。 最后, 发明描述了对裂缝进行判 定的方法, 统计裂缝面积大小, 以满足 PCI统计的要求。
红外图像处理: 红外图像处理分为两大步骤, 首先灰度化处理得到红外图像的灰度图, 然 后按照普通图像处理的步骤可以得到裂缝区域与路面区域 (图像中除裂缝的区域), 然后分别 计算两个区域的 RGB平均值, 与图例进行匹配得到两个区域的温度值, 进而计算裂缝与路面 的温度差数据。 如下图 4所示:
获取到红外图像后, 需要对图像进行处理得到沥青路面裂缝和路面的各自温度数据,然后 再进行进一步分析。 由红外热图像的灰度信息,利用现有的成熟算法可以得到裂缝区域和路面 区域, 然后求得各自区域内图像 RGB均值, 最后将平均的 RGB值与 colorbar进行匹配, 从左 向右依次进行,步进取两个像素宽度以保证温度值精度, 如图 5所示为红外图像处理技术及裂 缝与路面温差获取技术路线, 图 6为图像处理过程示意图。
在图像的采集过程中, 由于受到各种因素的影响, 没有经过处理的采集到的裂缝图像都会 包含有噪声信息。 如果不去除这些噪声, 会对后续的图像处理过程及结果分析带来困难。所以 图像的预处理是很有必要的,通过预处理可以把图像中的一些冗余信息去掉, 突出我们感兴趣 的目标, 从而达到减少图像信息量和改善图像质量的目的。 首先需要对图像进行降噪处理。
路面裂缝图像中需要识别的裂缝目标和背景相比较, 信息量比较少, 而且在图片的采集 和传输过程中, 由于很多因素的干扰也会使得图像的清晰度和对比度有所降低。 所以在对图 像进行滤波去噪之后, 还要对图形进一步增强, 使得我们感兴趣的裂缝目标更加突出, 为后 面的分割识别算法提供基础。
图像分割就是把图像中的像素进行分类, 分成一些特定的, 具有独特性质的区域并且提 取出感兴趣目标的技术和过程。 由于人们在对图像进行研究时, 都会有一个目标, 这就是图 像中人们感兴趣的区域, 而往往这些目标区域都有一些特定性质。 本发明此步骤目的就是要 找出裂缝区域和路面非裂缝区域。
( 1)温差麵
在完成裂缝区域和路面区域的区分后, 就可以根据各区域的平均 RGB值与 colorbar进行 匹配, 得到区域的温度值。 然后可以将温度数据进行处理。
如附图 7所示, 以温差为纵坐标、光照强度为横坐标作图, 其中温差选取大气温度大致相 同的情况下, 严重裂缝与路面的温度差。光照强度增大后会使路表的温度整体上升, 扩大路面 温度与大气温度的差别, 在实际实验中, 当气温为 35°C时, 光照强度为 15万 Lux时沥青路面 温度可以达到 55°C ; 但是, 照度对沥青路面和它表面的温度的影响几乎是同步的, 因此, 照度 对于裂缝和路面的温差基本没有影响。
因此, 在路面充分整洁和干燥的情况下, 我们通过查询相关资料将温差影响因素锁定在照 度和大气温度, 再通过实验, 分析出照度对裂缝与路面温差基本无影响。 需要注意的是, 实验 中有选取雨后的路面, 发现裂缝温度会比路面更低 (与完全干燥情况相反), 并且随着路面变 干过程中温差也会有较大变化,变化较复杂, 本发明通过将实验选取在完全干燥情况下避免裂 缝湿度的影响。
最后需要确定大气温度对裂缝与路面温差的影响关系。 由前文所述, 照度对于沥青路面裂 缝与路面的温差几乎无影响, 因此不考虑照度因素, 将所有不同温度下的温差数据进行绘图, 如附图 9所示。 不同发育程度的裂缝温差与大气温度敏感性大小不同, 裂缝发育程度越严重, 受大气温度影响相对越大,但总体趋势各个等级的裂缝与路面温差都随着大气温度的升高而变 大。
综上所述, 本发明只考虑裂缝与路面的温差和大气温度的关系。 同时由上图可知, 对于不 同发育程度的裂缝, 不同温度下其温差都是有差别的, 因此我们可以通过检测裂缝与路面的温 差来反映裂缝的发育程度。
关于图像分割:
传统基于图像分析的裂缝识别技术仅关注裂缝本身, 需要检测出裂缝的有无、位置和几何 形态等因素,裂缝与路面之间的过渡区域基本不作考虑。发明主要是要获取路面区域和裂缝区 域的温度差, 是需要关注裂缝和路面两个区域的, 因此可以考虑过渡区域的影响, 通过图像分 割获取裂缝和路面区域后, 需要对两个区域边界过渡区域进行删减, 因为路面温度到裂缝温度 是渐变的,裂缝和路面的过渡区温度介于裂缝中心温度和路面温度之间, 过渡区对于路面温度 的获取就是干扰区, 如图 10所示, 可以采取下列措施对干扰区进行排除。
对于路面区域, 过渡干扰区相对路面区域面积而言很小, 因此在进行排除的时候可以采取 固定的宽度进行路面区域的删减, 此处的宽度可以设定为 w宽 (0.5cm<w<2.5cm ), 实际还可 以采取基于裂缝区域最大宽度比值的方法, 即检测出裂缝区域后, 将裂缝区域上下各增加一倍 宽度后, 剩下的区域定为无干扰区的路面区域, 假设图像分割出来的路面区域宽度为 ¾, 最 终采取确定的宽度为 D , 则有以下公式 (6), 此方法可以较好的排除过渡干扰区的影响。
D = D0 _ 2d0 ( 6) 对于裂缝区域, 因为其宽度较小, 因此过渡干扰区的排除需要更细致, 可以采用基于裂缝 区域宽度比例的方法来进行排除,对图像分割的裂缝区域需要进行瘦身,假设裂缝的宽度为 图像分割出来的裂缝区域宽度为 dQ , 则满足以下条件如式 (7) 所示: d, = dn - lad^ ( 7)
«为为了排除过渡干扰区而对裂缝宽度上下部分分别删减的比例, 0.1≤ ≤0.2 . 过渡干扰区范围确定后, 如何具体对其进行删除可以参考以下: 对于路面区域, 相当于将 图像分割边界沿着分割边界径向往路面区域方向移动 距离得到新的路面区域边界, 对于裂 缝区域, 相当于将图像分割边界沿着分割边界径向往裂缝区域方向移动《 距离, 得到新的裂 缝区域边界。
关于区域 值计算:
沥青路面是指在矿质材料中掺入路用沥青材料铺筑的各种类型的路面。沥青结合料提高了 铺路用粒料抵抗行车和自然因素对路面损害的能力, 使路面平整少尘、 不透水、 经久耐用, 道 路结构是修筑在自然界中的带状结构物,环境因素和荷载的作用是造成路面结构破坏的主要原 因, 路面区域物理特性比较一致, 因此在红外图像中, 路面区域的温度值基本一样, 因此处理 起来比较方便, 不需要进行分别考虑, 即在图像分割中得到路面区域后就返回到红外图像中计 算该区域整体的 RGB平均值。
但是, 裂缝区域通常为细长的形状, 譬如一条两米长的裂缝, 可能只有几个毫米宽, 这种 情况下, 同一裂缝不同位置处的严重程度可能会有较大的差异, 两米长的裂缝有半米长的区域 发育程度比较严重, 根据平均 RGB值计算出来可能需要修补, 但是对于另外一条三米长的裂 缝, 其也有半米长的区域发育程度比较严重, 由平均 RGB值计算出来的温差得到的发育程度 指数就会比之前一条要低, 甚至显示为不修补, 显然是不科学的, 因此, 对于裂缝区域需要分 区域进行 RGB值得计算, 并对发育程度较严重的区域给予较高的权重, 这样计算出来一条裂 缝区域与路面的温差就会是一个数组, 如图 11所示, 不同区段裂缝可以得到不同温差值。
对于横向裂缝, 利用图像沿 y轴方向直线将有效裂缝区域分成所述 p段, 每一段长度任 意; 对于纵向裂缝, 利用图像沿 X轴方向直线将有效裂缝区域分成所述 p段, 每一段长度任 意; 对于其他种类裂缝, 不分段或者根据裂缝几何中心按角度分成所述 p段, 每一段对应中心 角度任意。 根据精度要求, 可以将图像分割出来的裂缝区域分成《(«≥2)段来进行考虑, 每一小段均 可以按照前文所述的技术路线进行处理, 即首先进行图像灰度化、 降噪, 然后进行图像增强后 再进行图像分割,得到裂缝区域后再分成《段进行后续处理,包括对每一段进行平均 RGB值得 计算, 再由平均 RGB值与图例中的颜色值进行匹配, 确定该段的温度值, 最后可以得到《段裂 缝区域的长度 /和裂缝区域与路面温度的温度差 ΔΤ, 即得到如下数组, 如式 (8) 所示: l ={h k … 0 ( 8) 其中 /, 即为裂缝区域划分区段后第 / 段的长度, 计算温度差数组如式 (9) 所示。 Γ = (ΑΤ1 ΑΤ2 ··· Δ7 ) (9) 其中 Δ7 即为裂缝区域划分区段后第 / 段与路面的温度差。 然后可以根据下式(10)进行加权平均计算, 得到所述有效路面区域和所述有效裂缝区域 的最终实测温差 Δ71 .
Ar = (10) 最后将此 Δ 与参照温差数据进行对比, 得到裂缝发育程度相关参数。
或者由 Δ 数组得到各段裂缝区域的发育程度指数如下式 (11): m = {n 2 … ") (11) 其中^ 即为裂缝区域划分区段后第 /段的裂缝发育程度指数。 根据裂缝的发育程度赋予不同区段的裂缝权重,因此可以得到裂缝的最终发育程度指数如 下式 (12): m = mlll+m2l2+- + mnln (丄
Ιλ +/2 +"' + /„ 计算结果四舍五入取 1、 2或 3, 即裂缝发育程度的三个等级。 发明中可以采取平均分配的方法,那么就有 ^二^-^ ,公式就可以化成下列形式(13): m =— (13)
n 关于建模 中裂缝发育程度指数的计算 在模型建立过程中, 发明通过机器学习将裂缝发育程度分为了 1、 2和 3三个等级, SVM 支持向量机, 作为一种分类方法, 通过核函数将低维空间上线性不可分的样本映射到高维空 间上线性可分的样本空间, 通过核函数计算内积, 得到一个线性分类器。 常用的核函数有多 种, 如线性核函数, 多项式核函数, 径向基核函数, Sigmoid核函数和复合核函数。 发明中 就采用线性核函数首先对 1、 2等级进行分类, 得到分类函数 /12 , 再根据数据对 2、 3等级进 行分类, 得到分类函数 /23, 在进行机器学习分类中会存在一定误差, 就分类采用数据中也会 有一部分属于较轻发育程度的裂缝被分配到了较严重发育程度的类别, 相反也会有较重发育 程度的裂缝分配到了较轻的类别。 但是, 在赋予权值的平均发育程度指数的计算中会出现有 小数的裂缝发育程度指数, 因为本身的分级就存在一定误差, 这些有小数的发育程度指数虽 然不一定可以使裂缝发育程度指数更精确, 但是却可以反映出相对大小, 2. 6的发育程度指 数四舍五入被划分为发育程度指数为 3的裂缝, 2. 4的就被划分为发育程度指数为 2的裂 缝, 这是之前的分类方法, 但是由赋予权重计算出来的 2. 6发育程度指数和 2. 4发育程度指 数实际上是差别不大的, 这对于管理者的决策有相当重要的影响。 因此, 在裂缝发育程度指 数的计算中, 发明可以取一位小数来表明计算出来的相对关系。 但是, 需要注意的是 2. 4的 发育程度指数不一定比 2. 3的发育程度指数要更严重, 因为前文说过, 本身分类就存在误 差, 2这个整数部分可能就已经是不准确的了。 因此, 为了避免这种问题, 发明在计算过程 中根据加权平均计算方法的方差确定了一个误差项, 即发育程度指数可以取一位小数并表示 为: m± 的形式。 采用支持向量机对数据进行线性分类。 以大气温度为横坐标、 裂缝与路面温度差为纵坐 标绘制点, 共有 1、 2、 3三个等级, 数字越大发育程度越严重, 可以得到如下两个分类函 数。
3、 2分类函数: ΔΓ23 = 3Γ+ 3 ( 14) 其中 ( °C ) 为大气温度, Γ CO 为沥青路面与裂缝的温度差; α23为线性分类函数 系数, 取值范围为 0.02~0.03, 3线性分类函数常数项, 取值范围为 0.60~0.85.
2、 1分类函数: ΔΓ12 = α12Γ+612 ( 15) 其中 ( °C ) 为大气温度, Γ CO 为沥青路面与裂缝的温度差; 12为线性分类函数 系数, 取值范围为 0.0075~0.0100, 2线性分类函数常数项, 取值范围为 0.4~0.65. 检测结果根据以下判断 首先根据大气温度计算出 ΔΓ12和 ΔΓ23, 然后将测量出的 Δ 与 ΔΓ12、 ΔΓ23进行对比, ΔΓ = ΔΓ12则发育程度为 1; ΔΓ = ΔΓ23则发育程度为 2, 介于之间的采用线性插值法计算发育 程度指数, 指数大于 3的取 3.
若要对裂缝区域进行区段划分, 并根据裂缝的发育程度赋予不同的权重,那么由公式计算 出来的结果可以取一位小数。
误差项 σ的取值可以由以下公式 (16) 得到:
Figure imgf000019_0001
Μ' 为对同一裂缝进行多次检测的结果, 然后由以上公式得出标准差, 据此可以将裂缝的 发育程度指数写到小数点后一位, 并通过一个固定的误差项来进行误差限定, 固定的误差项由 实验得到, 新的发育程度指数写法如下式 (17) :
m = m±a ( 17) 其中 "为误差项系数, 取值范围为广 3.
关于图像麵采集:
图像采集可以直接进行两路视频拍摄, 然后由图像处理技术直接对视频流进行实时处理, 得到道路的病害信息, 并参照时空坐标。
以上图像采集方式对于处理器的性能要求较高, 图像处理算法较复杂,但是对于设备的要 求较少, 不需要精确测定移动平台的速度。另外一种图像采集方式需要由光电编码器和同步控 制器进行配合,光电编码器精确获取移动平台的速度, 同步控制器由平台速度来确定每隔一定 距离进行图像拍摄, 进一步同步触发两台摄像机进行图像采集, 然后图数据传输到处理端, 处 理端直接对每一张图片进行处理,得到病害数据并将有病害的图片进行存储。此种采集方式可 以降低图像算法和对处理器的要求, 即通过增加硬件降低算法的复杂度。
光电编码器: 光电编码器, 是一种通过光电转换将输出轴上的机械几何位移量转换成脉冲 或数字量的传感器, 是目前应用最多的传感器。一般的光电编码器主要由光栅盘和光电探测装 置组成。 在伺服系统中, 由于光电码盘与电动机同轴, 电动机旋转时, 光栅盘与电动机同速旋 转. 经发光二极管等电子元件组成的检测装置检测输出若干脉冲信号。通过计算每秒光电编码 器输出脉冲的个数就能反映当前电动机的转速。此外, 为判断旋转方向, 码盘还可提供相位相 差 90° 的 2个通道的光码输出,根据双通道光码的状态变化确定电机的转向。根据检测原理, 编码器可分为光学式、 磁式、 感应式和电容式。 根据其刻度方法及信号输出形式, 可分为增量 式、 绝对式以及混合式 3种。
同步控制器: 同步控制器, 是指要按照一定比率来协调主机和从机之间的位置、 转速、 扭 矩等量, 同步控制器一般有两类。 一类是和张力系统连同一起来使用的, 象张力控制器也是一 种同步控制器件,这类型的同步是以转速和扭矩等量的同步来实现的; 另一类是空间定位控制 器, 就是位置同步, 一般应用于机器人, 数控机床, 飞剪等系统的轴间联动使用, 是一种轴间 的位置跟踪定位。 目前同步控制器有嵌入式设定参数的, 也有直接可编程类的, 随着技术的发 展, 可编程类的应用慢慢超过了前者, 代表者同步技术的发展方向, 它可以通过现场总线等通 讯技术和其他设备进行连接和操作。
关于数据采 境:
利用路面和裂缝的温度信息进行裂缝发育程度的检测, 而路面温度变化较大, 容易受环境 因素干扰,对实验环境的要求要比根据普通图像进行裂缝识别更高。我们根据路面温度变化的 规律来确定实验对环境的要求, 环境要求条件保证了最终模型建立的可靠性。
一天之中 , 路表面的最高温度几乎与日最高气温同时出现.在每天的日出和日落时段, 路 面内部各深度的温度总是大致相等.路表面的最低温度也几乎与日最低气温同时出现.同时, 随 着深度的增加, 在不同深度出现日最低温度的时间也逐渐滞后, 即太阳辐射虽然具有与路面温 度相近的变化规律, 但对路面温度的影响则呈现出滞后性和累积性的特点。 可以总结如下:
( 1) 晴朗的白天。 由于路面温度多变,容易受气候条件干扰, 因此实验尽量保证环境因 素稳定单一, 本实验选取实验地区常见的晴朗的白天。 夜间裂缝与路面温度差会较小, 如图所 示, 考虑是因为沥青普通温度变化规律所致。
( 2) 有效数据采集时间为上午八点至下午四点。 由上文分析, 日落和日出时段, 路面内 部各深度的温度总是大致相等, 因此此段时间内路面与裂缝温差很小, 通过实验也得到验证。
( 3 ) 路面完全干燥。雨后路面与裂缝都被水浸润,温度基本相同; 湿润的路面会因为水 的蒸发等使温度不稳定, 检测结果不稳定; 路面干燥而裂缝湿润, 水会影响裂缝的温度, 实验 中发现此情境下裂缝温度低于路面温度。
(4) 路面干净。为了保证裂缝和路面的红外线能顺利被热像仪检测,需要保证路面的干 净, 泥土、 杂质和油污等都会有影响。 附图简要说明 图 1 模型建立系统采样数据, 其中包括红外图像 (101 )、 大气温度 (102)、 裂缝发育程度 ( 103)、 实测温差数据 (104); 图 2 采集设备固定装置示意图;
图 3 采集端与处理层示意图;
图 4 红外图像温差示意图;
图 5 图像处理技术路线图;
图 6 图像处理过程图;
图 7 照度与温差关系图;
图 8 大气温度与温差关系图;
图 9 支持向量机线性分类图;
图 10 沥青路面裂缝区域、 路面区域和过渡干扰区;
图 11 不同裂缝区段获得不同温差示意图;
图 12 应用阶段采样数据, 其中包括红外图像 (201 )、 大气温度 (202 )、 普通图像 (203 )、 振动信息 (204)、 时空坐标 TS^ TS2和 TS3。 具体实鮮式
( 1) 模型建立
考虑到不同地区和不同季节的气候条件差异较大,需要进行裂缝发育程度检测模型的差异 化构建。 首先需要在合适的采集环境下进行沥青路面裂缝的红外图像采集, 并达到一定数量, 然后根据大气温度、路面与裂缝温差数据以及裂缝的发育程度进行支持向量机的分类学习,得 到分类函数, 确定此地区某季节的裂缝发育程度检测模型。
( 2) 环境确定
预测关系模型需要在一定环境条件下使用才能保证精确性, 首先需要保证数据采集时环 境条件稳定, 实验环境可以需要满足: 晴朗的白天、 有效数据采集时间为上午八点至下午四 点、 路面完全干燥和干净。
( 2) 数据采集
将普通相机(具有夜视功能)和红外相机(测温型) 固定在移动平台上, 在满足采集环境 下进行道路巡检,首先选取特定的车道,进行初始车道信息记录,巡检速度最高可达 80km/h, 在巡检过程中两路相机进行实时图像采集, 频率大于等于 60Hz, 振动传感器采集频率至少 100Hz, 沿线采集设备进行实时高频采样, 同时定位设备和系统时间实时进行时空坐标采集。 使用场景为各等级公路的沥青路面。 此阶段属于应用阶段, 具体的采样数据如图 12所示。
( 3 ) 数据处理
首先进行两路图像的实时处理,对每一张进行路面病害识别算法处理, 并对图片进行有无 病害的分类存储,对于有病害的图像进一步识别病害的类型, 对于裂缝型病害进一步根据红外 图像进行裂缝发育程度的识别, 然后将病害形成日志信息, 与定位信息进行匹配, 同一裂缝可 能存在多个图像中, 处理后每个图像都会得到病害信息, 在与定位信息进行匹配后则可以确定 病害的唯一性, 即同一位置坐标处的同一病害类型只存在一个。最后将病害信息上传至中心服 务器, 待下一步展示和应用。
同时进行振动信号的处理,通过高频振动信号实时计算路段的平整度信息, 并与时空坐标 进行匹配后形成完整的平整度检测信息后上传至服务器数据库待下一步展示和应用。
(4) 数据发布
数据发布有两个层次, 第一层是作业层的展示发布, 在车载端配置数据发布层, 显示端直 接与处理器连接,数据来源于实时的处理端传输信息, 主要是对数据流的可视化显示以及检测 工况的工作状态显示, 无交互界面。 工况分为正常、 异常和错误, 数据流的显示辅助相关人员 进行异常工况的排除, 错误工况属于设备问题, 需要进行停机检查。 异常工况主要是采集气候 条件不满足或者移动平台速度过快等异常采集环境条件。
第二层是应用层的发布, 主要是可视化 Web形式, 分为日志界面和可视化地图界面, 日 志页面可以实现数据的查询导出导入等操作,可视化界面利用在二维地图上显示病害的位置和 信息, 同时可以切换至平整度显示模式。

Claims

权利要求书
1. 一种基于双目图像分析的沥青路面病害检测系统, 包括裂缝发育程度检测模 型建立、 信息采集、 信息分析、 信息传输和信息发布五个子系统-
1) 模型建立子系统- a. 采集沥青路面至少 10个样本裂缝区的红外图像(101), 同时记录样本大 气温度 (102)及所述样本裂缝区内的裂缝发育程度 (103), 采用传统裂缝 严重程度分类方法对裂缝发育程度进行估计, 将所采集裂缝的裂缝发育程 度 (103) 分为 1、 2、 3三个等级;
b. 对步骤 a中裂缝区的红外图像(101)分别进行处理, 得到对应的裂缝区 内的裂缝和裂缝区内的路面之间的实测温差数据 ( 104);
c 利用步骤 b所得到的实测温差数据 (104), 以及步骤 a所记录的样本大 气温度(102)和裂缝区内的样本裂缝发育程度数据(103)进行支持向量机 分类,得到两个分类函数 Δ ^7^ 和 Δ =«23^23 ,所述分类函数自 变量为所记录的样本大气温度 Γ, 因变量为参照温差数据 Δ7 ^ΡΔΓ23;
2)采集子系统,所述采集子系统位于移动平台上,移动平台在路段上持续移动, 所有采集行为都是连续的:
d. 采集沥青路面裂缝区的红外图像 (201), 同时记录大气温度 (202)、 采 集时空坐标 TS1;
e. 采集沥青路面的普通图像 (203)、 同时采集时空坐标 TS2;
f. 采集移动平台后轴振动信息 (204)、 同时采集时空坐标 TS3;
3)分析子系统,所述分析子系统内嵌于处理端,处理端的设备位于移动平台上: h. 将 d、 e步骤中的红外图像 (201)、 大气温度 (202) 和普通图像 (203) 进行处理, 并与步骤 e中的时空坐标 TS TSz匹配, 得到病害信息; i. 将 f步骤中的振动信息 (204) 进行处理, 并与步骤 e中的时空坐标 TS3 匹配, 得到平整度信息;
4) 传输子系统:
j. 将步骤 h、 i中的病害信息和平整度信息上传至服务器数据库;
5) 发布子系统:
k. 利用步骤 j中服务器数据库的病害信息和平整度信息建立可视化界面。
2. 如权利要求 1所述的沥青路面病害检测系统, 其特征在于, 步骤 d包括: 利用非制冷焦平面测温型红外热像仪拍摄沥青路面红外图像 (201), 设 备距离地面 1至 2m;
d2. 利用气温计在采集上述红外图像 (201) 的同时记录大气温度 (202)。
3. 如权利要求 1所述的沥青路面病害检测系统, 其特征在于, 步骤 e包括: d. 采用高清摄像机拍摄沥青路面普通图像 (203), 设备距离地面 1至 2m。
4. 如权利要求 1所述的沥青路面病害检测系统, 其特征在于, 步骤 f包括: fi.将两个采集频率范围在 20Hz〜500Hz之间的传感器模块分别固定在车辆后 轴上车轮的正上方;
f2.在移动平台启动后以特定频率开始采集平台竖直方向的振动信息 (204); f3.得到振动信息 (204) 后通过无线传输方式输出至处理端。
5. 如权利要求 1所述的沥青路面病害检测系统, 其特征在于, 步骤 d、 e、 f 中 所采集的:
时间坐标统一由处理端时间确定;
空间坐标由定位设备采集, 时空坐标 TS^ TS2、 TS3在处理端进行保存。
6. 如权利要求 1所述的沥青路面病害检测系统, 其特征在于, 步骤 h包括: h,将普通图像(201)利用病害识别模型算法进行处理得到病害有无和类型; W 将红外图像 (203) 利用病害识别模型算法进行处理, 得到裂缝区域; h3. 将红外图像 (203) 利用裂缝发育程度检测模型进行处理, 得到裂缝的发 育程度;
h4. 将上述裂缝有无、 类型和发育程度信息, 贴上时空坐标 TS TS2标签。
7. 如权利要求 6所述的沥青路面病害检测系统, 其特征在于, 步骤 h3包括: h3,对于所述沥青路面红外图像(201)进行图例匹配,得到路面实测温度 Γ。; h32.将步骤 h2经过图像处理得到的所述裂缝区域划分为 ρ段, p^2, 每段长 度为! ·!Ρ, 分别对每一段进行图例匹配, 得到实测温度 … ρ ; h33.将步骤 h2中所述裂缝区域各段实测温度减去步骤 h31所述的路面实测温度 T0, 得到实测温差 Δ7
h34. 根据下式进行加权平均计算, 得到最终实测温差 ΔΓ,
h .将大气温度 (202) 带入所述检测模型中的分类函数 Δ 1212 + 12, 得到 参照温差数据八72; h36.将所述的大气温度(202)带入所述检测模型中的分类函数 Δ 23 =a13T+b13, 得到参照温差数据 Δ 23 ;
h37.对步骤 h 34中的 ΔΓ、 h 中的 ΔΓ12、和 h 36中的 ΔΓ23进行比较,若 ΔΓ < ΔΓ12, 发育程度为 1 ; 若 ΔΤ12 < ΔΤ < ΔΤ23, 发育程度为 2; 若 ΔΤ〉ΔΤ23, 发育程度 为 3。
8. 如权利要求 1至 7之一所述的沥青路面病害检测系统, 其特征在于, 步骤 i 包括:
Ϊ!. 将振动信息(204)利用基于功率谱密度的平整度评估模型算法进行处理, 得到路段的平整度信息;
. 将 所述平整度信息,贴上时空坐标 TS3标签,其中时间坐标精确到分钟, 空间坐标精确到米。
9. 如权利要求 1至 7之一所述的沥青路面病害检测系统, 其特征在于, 步骤 k 包括:
将所述保存在服务器数据库的病害信息和平整度信息通过类别、 时间序列 日志形式在可视化界面展示;
k2. 将所述保存在服务器数据库的病害信息和平整度信息通过类别、 空间序列 进行地图可视化。
10.如权利要求 9所述的沥青路面病害检测系统, 其特征在于:
步骤 ^中的:
所述日志形式可以分类査询、 按时间和空间进行排序;
所述日志形式可以导入外部病害信息并进行保存、 调用;
所述日志支持报表生成, 具有统计图表可视化功能, 并支持分析模型接入; 步骤 k2中的:
所述地图可视化可以分类査看、 按空间进行排序;
所述地图可视化支持单点单病害査询, 并生成报表。
PCT/IB2017/058549 2016-12-30 2017-12-30 一种基于双目图像分析的沥青路面病害检测系统 Ceased WO2018122819A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CN201780056387.4A CN109716108B (zh) 2016-12-30 2017-12-30 一种基于双目图像分析的沥青路面病害检测系统
US16/474,710 US11486548B2 (en) 2016-12-30 2017-12-30 System for detecting crack growth of asphalt pavement based on binocular image analysis
GB1905912.0A GB2571016B (en) 2016-12-30 2017-12-30 A system for detecting crack growth of asphalt pavement based on binocular image analysis

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
PCT/IB2016/058109 WO2018122589A1 (zh) 2016-12-30 2016-12-30 一种基于红外热像图分析的沥青路面裂缝发育程度检测方法
IBPCT/IB2016/058109 2016-12-30

Publications (1)

Publication Number Publication Date
WO2018122819A1 true WO2018122819A1 (zh) 2018-07-05

Family

ID=59713564

Family Applications (5)

Application Number Title Priority Date Filing Date
PCT/IB2016/058109 Ceased WO2018122589A1 (zh) 2016-12-30 2016-12-30 一种基于红外热像图分析的沥青路面裂缝发育程度检测方法
PCT/IB2017/058540 Ceased WO2018122810A1 (zh) 2016-12-30 2017-12-30 一种基于动态红外热像图处理的地下管廊渗漏检测方法
PCT/IB2017/058539 Ceased WO2018122809A1 (zh) 2016-12-30 2017-12-30 一种基于静态红外热像图处理的地下管廊渗漏检测方法
PCT/IB2017/058549 Ceased WO2018122819A1 (zh) 2016-12-30 2017-12-30 一种基于双目图像分析的沥青路面病害检测系统
PCT/IB2017/058548 Ceased WO2018122818A1 (zh) 2016-12-30 2017-12-30 一种基于红外热像图分析的沥青路面裂缝发育程度检测方法

Family Applications Before (3)

Application Number Title Priority Date Filing Date
PCT/IB2016/058109 Ceased WO2018122589A1 (zh) 2016-12-30 2016-12-30 一种基于红外热像图分析的沥青路面裂缝发育程度检测方法
PCT/IB2017/058540 Ceased WO2018122810A1 (zh) 2016-12-30 2017-12-30 一种基于动态红外热像图处理的地下管廊渗漏检测方法
PCT/IB2017/058539 Ceased WO2018122809A1 (zh) 2016-12-30 2017-12-30 一种基于静态红外热像图处理的地下管廊渗漏检测方法

Family Applications After (1)

Application Number Title Priority Date Filing Date
PCT/IB2017/058548 Ceased WO2018122818A1 (zh) 2016-12-30 2017-12-30 一种基于红外热像图分析的沥青路面裂缝发育程度检测方法

Country Status (4)

Country Link
US (2) US11486548B2 (zh)
CN (5) CN109804119B (zh)
GB (6) GB201711412D0 (zh)
WO (5) WO2018122589A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109903325A (zh) * 2019-04-03 2019-06-18 杭州晶一智能科技有限公司 基于立体视觉深度信息的地面精确描述方法
CN109919139A (zh) * 2019-04-01 2019-06-21 杭州晶一智能科技有限公司 基于双目立体视觉的路面状况快速检测方法
CN115272222A (zh) * 2022-07-25 2022-11-01 中科海微(北京)科技有限公司 道路检测信息的处理方法、装置、设备及存储介质

Families Citing this family (239)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10970590B2 (en) * 2015-06-05 2021-04-06 Schlumberger Technology Corporation Image-based wellsite equipment health monitoring
US12276420B2 (en) 2016-02-03 2025-04-15 Strong Force Iot Portfolio 2016, Llc Industrial internet of things smart heating systems and methods that produce and use hydrogen fuel
US10551297B2 (en) * 2017-09-22 2020-02-04 Saudi Arabian Oil Company Thermography image processing with neural networks to identify corrosion under insulation (CUI)
CN107767377B (zh) * 2017-11-09 2024-02-06 高视科技(苏州)股份有限公司 基于双目视觉系统的液晶屏缺陷与灰尘区分法及检测装置
CN108364280B (zh) * 2018-01-03 2022-04-15 东南大学 结构裂缝自动化描绘及宽度精准测量方法与设备
US20200133254A1 (en) 2018-05-07 2020-04-30 Strong Force Iot Portfolio 2016, Llc Methods and systems for data collection, learning, and streaming of machine signals for part identification and operating characteristics determination using the industrial internet of things
CN109058770B (zh) * 2018-07-17 2019-06-07 北京讯腾智慧科技股份有限公司 一种基于北斗差分定位的燃气管道的修补方法及系统
US10643324B2 (en) 2018-08-30 2020-05-05 Saudi Arabian Oil Company Machine learning system and data fusion for optimization of deployment conditions for detection of corrosion under insulation
US10533937B1 (en) 2018-08-30 2020-01-14 Saudi Arabian Oil Company Cloud-based machine learning system and data fusion for the prediction and detection of corrosion under insulation
JP7056804B2 (ja) * 2018-09-28 2022-04-19 日本電気株式会社 経験損失推定システム、経験損失推定方法および経験損失推定プログラム
CN109488888B (zh) * 2018-11-06 2020-07-17 沈阳天眼智云信息科技有限公司 基于对红外温度场多元分析的金属管道泄漏监测方法
CN109358060B (zh) * 2018-11-14 2023-08-25 浙江工业大学 一种用于检测并标记道路地砖松动情况的系统及方法
US10878596B2 (en) 2018-11-29 2020-12-29 International Business Machines Corporation Object oriented image normalization
US11143599B2 (en) 2018-12-03 2021-10-12 Mistras Group, Inc. Systems and methods for inspecting pipelines using a pipeline inspection robot
US10783623B2 (en) 2018-12-03 2020-09-22 Mistras Group, Inc. Systems and methods for inspecting pipelines using a robotic imaging system
CN109632963B (zh) * 2019-01-11 2020-10-20 南京航空航天大学 一种基于时不变特征信号构建的结构损伤四维成像方法
EP3909223B1 (en) 2019-01-13 2024-08-21 Strong Force Iot Portfolio 2016, LLC Monitoring and managing industrial settings
CN109859125B (zh) * 2019-01-14 2022-10-21 广东工业大学 基于形态学检测与小波变换的图像高光修复方法
CN109801282A (zh) * 2019-01-24 2019-05-24 湖北大学 路面状况检测方法、处理方法、装置及系统
CN109816656B (zh) * 2019-02-01 2023-06-20 辽宁工程技术大学 一种火电厂负压侧系统漏点精确定位方法
CN109800739B (zh) * 2019-02-21 2023-04-25 四川中天炬矿业有限公司 电加热旋转窑温度检测装置
CN109858570A (zh) * 2019-03-08 2019-06-07 京东方科技集团股份有限公司 图像分类方法及系统、计算机设备及介质
CN110232666B (zh) * 2019-06-17 2020-04-28 中国矿业大学(北京) 基于暗原色先验的地下管道图像快速去雾方法
CN110349134B (zh) * 2019-06-27 2022-12-09 广东技术师范大学天河学院 一种基于多标签卷积神经网络的管道病害图像分类方法
CN110205910A (zh) * 2019-07-08 2019-09-06 河北工程大学 一种铺面结构表面平整性智能检测与分析装置
CN110320236B (zh) * 2019-07-19 2021-09-14 沈阳工业大学 大型风力机叶片内部缺陷深度的红外测量方法
CN110490862B (zh) * 2019-08-22 2022-08-09 联峰钢铁(张家港)有限公司 一种提高连铸探伤合格率的方法及装置
JP7285174B2 (ja) * 2019-09-04 2023-06-01 株式会社トプコン 壁面のひび割れ測定機および測定方法
CN110674754A (zh) * 2019-09-25 2020-01-10 武汉易视维科技有限公司 一种在线间歇中空滤棒视觉缺陷检测识别系统
CN110687277B (zh) * 2019-10-28 2025-05-02 上海工程技术大学 一种轨道板温度裂缝模拟装置及其实现方法
CN111025285B (zh) * 2019-11-01 2021-08-24 长安大学 一种基于图谱灰度自适应选取的沥青路面水损害检测方法
CN110793722B (zh) * 2019-11-08 2021-11-30 国家计算机网络与信息安全管理中心 基于机器学习的铅酸蓄电池非接触式漏液检测装置及方法
CN110763409B (zh) * 2019-11-27 2025-02-28 江苏省建筑工程质量检测中心有限公司 一种综合管廊拼缝密封性检测方法及装置
CN111062959B (zh) * 2019-11-28 2022-04-12 重庆大学 一种航空薄壁微构零件切出底边毛刺特征的提取及表征方法
CN111209794A (zh) * 2019-12-11 2020-05-29 浙江省交通运输科学研究院 一种基于探地雷达图像的地下管道识别方法
CN111191540A (zh) * 2019-12-20 2020-05-22 数海信息技术有限公司 一种基于温度梯度的对象状态分析方法及系统
US11651278B2 (en) 2019-12-23 2023-05-16 Saudi Arabian Oil Company Pipeline sensor integration for product mapping
CN111145205B (zh) * 2019-12-24 2022-04-12 天津农学院 基于红外图像的多猪只场景下猪体温检测方法
CN111210405A (zh) * 2019-12-27 2020-05-29 山东省特种设备检验研究院有限公司 一种基于红外成像的挥发性有机物泄露检测方法
US11774990B2 (en) 2019-12-30 2023-10-03 Marathon Petroleum Company Lp Methods and systems for inline mixing of hydrocarbon liquids based on density or gravity
US11247184B2 (en) 2019-12-30 2022-02-15 Marathon Petroleum Company Lp Methods and systems for spillback control of in-line mixing of hydrocarbon liquids
US11607654B2 (en) 2019-12-30 2023-03-21 Marathon Petroleum Company Lp Methods and systems for in-line mixing of hydrocarbon liquids
CN111242123B (zh) * 2020-01-07 2022-10-28 西安交通大学 一种基于红外图像的电力设备故障诊断方法
CN111209876B (zh) * 2020-01-10 2023-04-07 汕头大学 一种漏油缺陷检测方法及系统
US11054538B1 (en) * 2020-01-27 2021-07-06 S-L Ip, Llc Modification and assessment
CN111400891B (zh) * 2020-03-11 2023-08-11 中煤航测遥感集团有限公司 管道巡检点偏差程度获取方法、装置、设备及存储介质
CN111583413B (zh) * 2020-03-24 2023-06-02 交通运输部科学研究院 一种路面病害bim参数化建模与增强现实移动巡检方法
CN111445462A (zh) * 2020-03-30 2020-07-24 国家计算机网络与信息安全管理中心 一种基于神经网络和热像图的蓄电池漏液检测方法
CN111524146B (zh) * 2020-04-13 2023-05-05 同济大学 基于红外图像的暖通空调热水管道故障诊断方法及系统
CN111489352B (zh) * 2020-04-29 2023-06-02 安徽国钜工程机械科技有限公司 一种基于数字图像处理的隧道缝隙检测与测量方法及装置
CN111583362A (zh) * 2020-04-30 2020-08-25 东南大学 一种沥青路面病害状况检测的可视化记录方法
US11645766B2 (en) 2020-05-04 2023-05-09 International Business Machines Corporation Dynamic sampling for object recognition
TW202200978A (zh) * 2020-05-08 2022-01-01 香港商維視拍智能檢測有限公司 用於具人工智慧動力之檢驗及預測分析的系統及方法
CN111754455B (zh) * 2020-05-15 2024-03-19 华能国际电力股份有限公司海门电厂 一种火电厂漏水检测方法和系统
CN111640117B (zh) * 2020-06-03 2024-03-05 四川正大新材料科技有限公司 一种寻找建筑物渗漏源位置的方法
CN111738990A (zh) * 2020-06-03 2020-10-02 东北林业大学 基于log算法的损伤水果温度场检测方法
CN111784645B (zh) * 2020-06-15 2023-08-22 北京科技大学 一种充填管道裂纹检测方法
CN111746537B (zh) * 2020-06-22 2022-05-17 重庆长安汽车股份有限公司 基于路面平整度的自适应巡航车速控制系统、方法及车辆
CN111767815A (zh) * 2020-06-22 2020-10-13 浙江省机电设计研究院有限公司 一种隧道渗漏水识别方法
CN111693536A (zh) * 2020-06-24 2020-09-22 河南高建工程管理有限公司 一种基于红外热成像的地下综合管廊施工防水检测方法
CN111750283A (zh) * 2020-06-26 2020-10-09 西北工业大学 基于深度学习的强背景噪声环境下的气体管道泄漏识别方法
CN111795883B (zh) * 2020-07-06 2023-04-25 中国石油天然气集团有限公司 一种含裂纹金相试样的浸蚀及图像处理方法和设备
CN111767874B (zh) * 2020-07-06 2024-02-13 中兴飞流信息科技有限公司 一种基于深度学习的路面病害检测方法
CN112036073A (zh) * 2020-07-16 2020-12-04 成都飞机工业(集团)有限责任公司 一种3d打印零件测量结果矫正方法
CN111812215B (zh) * 2020-07-22 2021-06-29 南京航空航天大学 一种飞行器结构损伤的监测方法
CN114002222B (zh) * 2020-07-28 2024-05-14 宝山钢铁股份有限公司 一种用于在役管线的智能探伤装置、系统及其方法
CN111811933B (zh) * 2020-07-31 2022-03-11 中国矿业大学 一种承载煤岩损伤破裂过程中的红外辐射信息去噪方法
CN112032567A (zh) * 2020-08-14 2020-12-04 常州机电职业技术学院 一种埋地燃气管道泄漏危险度预测系统
CN112032568A (zh) * 2020-08-14 2020-12-04 常州机电职业技术学院 一种埋地燃气管道泄漏危险度预测算法及预测方法
CN112004093B (zh) * 2020-09-02 2022-07-12 烟台艾睿光电科技有限公司 一种红外数据压缩方法、装置及设备
CN111999001B (zh) * 2020-09-09 2022-07-08 中国南方电网有限责任公司超高压输电公司大理局 一种基于图像处理的换流站空冷器泄漏检测方法及系统
CN112085037B (zh) * 2020-09-21 2024-04-09 国网吉林省电力有限公司电力科学研究院 一种变电设备红外热故障特征提取及数字化表达方法
CN112183516A (zh) * 2020-09-22 2021-01-05 广东中鹏热能科技有限公司 基于图形图像识别技术对单照片多区域最高温的读取方法
CN112367461B (zh) * 2020-10-27 2021-12-17 京东方科技集团股份有限公司 仪表图像样本制作方法及系统、存储介质、电子设备
CN112270663B (zh) * 2020-10-27 2023-11-24 北京京能东方建设工程有限公司 基于蜂巢网络环境的沥青路面过筛修复系统
CN112345084B (zh) * 2020-11-05 2021-09-28 北京易达恩能科技有限公司 基于数字孪生环境的三维温度场构建方法及装置
CN112347919B (zh) * 2020-11-06 2023-07-25 中国矿业大学(北京) 一种地下天然气微泄漏点的遥感探测方法
CN112381796B (zh) * 2020-11-16 2021-08-03 广东电网有限责任公司肇庆供电局 一种基于红外数据的导线缺陷识别及检测方法
CN112464779B (zh) * 2020-11-23 2022-10-11 武汉舜陈技术有限公司 一种基于柔性材料的图形识别定位方法
CN112485329B (zh) * 2020-11-27 2024-01-26 重庆商勤科技有限公司 基于热成像与超声相结合检测排污口的方法、装置及系统
CN112541887B (zh) * 2020-12-02 2024-05-03 中国华能集团有限公司南方分公司 一种火电厂多管道设备运行现场漏水缺陷检测方法
CN112698015B (zh) * 2020-12-08 2024-04-02 温州鼎玛建筑技术有限公司 一种道路桥梁裂缝检测系统
CN112417378A (zh) * 2020-12-10 2021-02-26 常州大学 一种基于无人机图像处理的中华绒螯蟹质量估算方法
CN112609547A (zh) * 2020-12-11 2021-04-06 中山火炬职业技术学院 一种沥青路面各施工阶段层厚的监测方法
CN112560707B (zh) * 2020-12-18 2022-10-21 中国民用航空总局第二研究所 基于激光光源的移动式道面检测方法及系统
CN112508944B (zh) * 2020-12-27 2024-09-27 中信重工开诚智能装备有限公司 一种应用于煤矿井下供水管路的泄漏检测方法
CN112766251B (zh) * 2020-12-30 2022-06-14 广东电网有限责任公司佛山供电局 变电设备红外检测方法、系统、储存介质及计算机设备
CN112767322B (zh) * 2021-01-05 2023-06-13 成都圭目机器人有限公司 一种机场水泥道面fod风险评估方法和装置
CN112881432B (zh) * 2021-01-12 2022-11-29 成都泓睿科技有限责任公司 一种带液玻璃瓶瓶口裂纹检测方法
CN112763349B (zh) * 2021-01-21 2021-11-26 北京航空航天大学 一种复合材料结构冲击损伤的监测方法
CN112834457B (zh) * 2021-01-23 2022-06-03 中北大学 基于反射式激光热成像的金属微裂纹三维表征系统及方法
CN112669316B (zh) * 2021-01-29 2023-05-30 南方电网调峰调频发电有限公司 电力生产异常监控方法、装置、计算机设备和存储介质
US20240167962A1 (en) * 2021-03-15 2024-05-23 Iris R&D Group Inc. System and method for automatic monitoring of pavement condition
US11578638B2 (en) 2021-03-16 2023-02-14 Marathon Petroleum Company Lp Scalable greenhouse gas capture systems and methods
US11578836B2 (en) 2021-03-16 2023-02-14 Marathon Petroleum Company Lp Scalable greenhouse gas capture systems and methods
US12012883B2 (en) 2021-03-16 2024-06-18 Marathon Petroleum Company Lp Systems and methods for backhaul transportation of liquefied gas and CO2 using liquefied gas carriers
US11655940B2 (en) 2021-03-16 2023-05-23 Marathon Petroleum Company Lp Systems and methods for transporting fuel and carbon dioxide in a dual fluid vessel
CN112927223A (zh) * 2021-03-29 2021-06-08 南通大学 一种基于红外热成像仪的玻璃幕墙检测方法
CN114333096B (zh) * 2021-04-09 2023-12-15 北京市燃气集团有限责任公司 一种巡检机器人
CN113203743B (zh) * 2021-05-20 2023-12-12 中铁二十一局集团第四工程有限公司 一种基于红外热成像分析的路基裂缝检测识别及修复方法
CN113252724B (zh) * 2021-05-21 2022-05-31 山东中坚工程质量检测有限公司 一种外墙保温性能的检测方法
CN113177611B (zh) * 2021-05-24 2022-11-01 河北工业大学 基于力学指标和人工神经网络的路面病害快速巡检方法
EP4357746A4 (en) * 2021-06-16 2024-10-30 Konica Minolta, Inc. DEVICE FOR ESTIMATING THE GAS CONCENTRATION FUNCTION, METHOD FOR ESTIMATING THE GAS CONCENTRATION FUNCTION, PROGRAM AND DEVICE FOR GENERATING A GAS CONCENTRATION FUNCTION INTERVENTION MODEL
CN113375065B (zh) * 2021-07-01 2022-05-24 北京化工大学 管道泄漏监测中趋势信号的消除方法及装置
DE102021207204A1 (de) * 2021-07-08 2023-01-12 Zf Friedrichshafen Ag System und Verfahren zum Schätzen der Tiefe mindestens eines zumindest zum Teil mit Wasser gefüllten Schlaglochs und entsprechendes Fahrerassistenzsystem
CN113628164B (zh) * 2021-07-12 2024-08-06 北京科技大学 一种基于深度学习与web端定位的路面裂缝检测方法
CN113606502B (zh) * 2021-07-16 2023-03-24 青岛新奥燃气设施开发有限公司 一种基于机器视觉判断操作人员执行管道漏气检测的方法
CN113592798B (zh) * 2021-07-21 2023-08-15 山东理工大学 一种道路病害智能辨识方法、系统、终端及介质
CN113849901B (zh) * 2021-07-28 2024-05-03 上海机电工程研究所 针对接触换热系数辨识的改进自适应优化方法及系统
CN113591721B (zh) * 2021-08-02 2022-01-25 山东省交通科学研究院 一种利用无人机确定新摊铺沥青路面取芯点位置的方法
US12129559B2 (en) 2021-08-26 2024-10-29 Marathon Petroleum Company Lp Test station assemblies for monitoring cathodic protection of structures and related methods
US12043905B2 (en) 2021-08-26 2024-07-23 Marathon Petroleum Company Lp Electrode watering assemblies and methods for maintaining cathodic monitoring of structures
US12180597B2 (en) 2021-08-26 2024-12-31 Marathon Petroleum Company Lp Test station assemblies for monitoring cathodic protection of structures and related methods
US11447877B1 (en) 2021-08-26 2022-09-20 Marathon Petroleum Company Lp Assemblies and methods for monitoring cathodic protection of structures
CN113838078B (zh) * 2021-09-06 2023-06-30 中国矿业大学(北京) 采煤塌陷地裂缝的识别与提取方法、装置及存储介质
CN113503974B (zh) * 2021-09-09 2021-11-23 江苏沃泰冶金设备有限公司 基于pid的热成像检测系统、方法及瓦斯灰输送装置
CN113963285B (zh) * 2021-09-09 2022-06-10 山东金宇信息科技集团有限公司 一种基于5g的道路养护方法及设备
CN113884464B (zh) * 2021-09-27 2024-04-26 西安空天能源动力智能制造研究院有限公司 一种基于红外热像仪的涂层波段发射率外场测量方法
CN113947240B (zh) * 2021-09-30 2024-08-27 西安理工大学 一种基于时间序列的用水量预测方法、系统及设备
CN115014220B (zh) * 2021-10-15 2025-03-14 广州城建职业学院 基于误差控制的裂纹不连续区的全场应变测量方法
CN114239194B (zh) * 2021-10-20 2024-07-30 中州水务控股有限公司 一种大水量输供水管网漏损分析和漏点定位方法
US11598689B1 (en) 2021-10-24 2023-03-07 Philip Becerra Method of detecting and identifying underground leaking pipes
CN114005031A (zh) * 2021-10-27 2022-02-01 赛轮集团股份有限公司 一种硫化地沟油气管道异常检测方法、装置、电子设备及存储介质
CN114021702A (zh) * 2021-11-04 2022-02-08 福建省华渔教育科技有限公司 一种基于cnn-lstm混合模型的塔吊裂缝生长预测方法
CN114034405B (zh) * 2021-11-08 2024-06-25 北京航空航天大学 一种非接触式测温方法及系统
CN113777028B (zh) * 2021-11-11 2022-01-18 成都理工大学 测量凝胶类堵漏材料与岩石壁面粘附强度的装置和方法
CN114113217B (zh) * 2021-11-15 2024-07-12 中国矿业大学 一种煤岩体损伤程度的红外辐射量化评价方法
CN114037633B (zh) * 2021-11-18 2022-07-15 南京智谱科技有限公司 一种红外图像处理的方法及装置
CN114049336B (zh) * 2021-11-18 2024-06-14 国网重庆市电力公司电力科学研究院 Gis套管温度异常检测方法、装置、设备及可读存储介质
CN114384073B (zh) * 2021-11-30 2023-08-04 杭州申昊科技股份有限公司 一种地铁隧道裂纹检测方法及系统
CN114170552A (zh) * 2021-12-09 2022-03-11 上海谱盟光电科技有限公司 基于红外热成像的天然气泄漏实时预警方法及系统
CN114677314B (zh) * 2021-12-17 2024-08-06 上海大学 一种基于多尺度特征对齐及精化网络的双目图像修复网络模型的训练方法
CN116356650A (zh) * 2021-12-28 2023-06-30 北京万集科技股份有限公司 路面平整度检测方法及装置
CN114487012B (zh) * 2021-12-29 2023-11-03 南京大学 一种土体表面裂隙发育预判方法
CN114383789B (zh) * 2022-01-05 2024-06-18 中国科学院合肥物质科学研究院 基于激励源与容性阻尼模型的金属容器气密性红外检测方法
CN114363520A (zh) * 2022-01-11 2022-04-15 河北德冠隆电子科技有限公司 一种自动巡视快速定位目标的方法
CN114577399B (zh) * 2022-01-18 2024-07-19 潍柴动力股份有限公司 发动机漏气检测方法及检测装置
CN114782734A (zh) * 2022-01-19 2022-07-22 岳阳高澜节能装备制造有限公司 一种阀冷系统管道渗漏视觉检测方法及计算机与存储介质
CN114119614B (zh) * 2022-01-27 2022-07-26 天津风霖物联网科技有限公司 一种远程检测建筑物的裂缝的方法
CN114459708A (zh) * 2022-02-10 2022-05-10 合肥永信科翔智能技术有限公司 一种基于气体智能感知的泄漏气体监测系统
TWI833168B (zh) * 2022-02-23 2024-02-21 南亞科技股份有限公司 異常診斷方法
CN114721369B (zh) * 2022-02-28 2024-09-20 上海电气集团股份有限公司 一种应用于刚性填埋场的智能巡检系统
CN114577405A (zh) * 2022-02-28 2022-06-03 辽宁石油化工大学 一种基于红外成像技术的城市地下污水管道缺陷检测装置
CN114565793B (zh) * 2022-02-28 2023-05-23 湖南北斗微芯产业发展有限公司 一种道路交通裂缝监测方法以及系统
CN114663408B (zh) * 2022-03-30 2023-04-28 江苏天晶智能装备有限公司 一种基于人工智能的门窗水密性能检测方法
CN114913316B (zh) * 2022-04-02 2023-04-07 淮沪电力有限公司田集第二发电厂 工业设备表计识别的图像分类方法、装置、电子设备和存储介质
CN114511469B (zh) * 2022-04-06 2022-06-21 江苏游隼微电子有限公司 一种图像智能降噪先验检测方法
CN114459728B (zh) * 2022-04-13 2022-06-24 中国空气动力研究与发展中心高速空气动力研究所 一种低温温敏漆转捩测量试验方法
CN114838297B (zh) * 2022-04-14 2024-03-15 七腾机器人有限公司 一种原油管道泄漏检测方法、装置、存储介质及系统
US11686070B1 (en) 2022-05-04 2023-06-27 Marathon Petroleum Company Lp Systems, methods, and controllers to enhance heavy equipment warning
CN114972757B (zh) * 2022-05-31 2024-08-09 山东大学 一种隧道渗漏水区域识别方法及系统
CN114972314B (zh) * 2022-06-22 2025-02-28 广东电网有限责任公司 电力设备的裂缝检测方法、计算机设备和存储介质
CN115060763B (zh) * 2022-07-05 2024-08-06 海腾创建(深圳)集团有限公司 一种基于人工智能的墙面漏水检测设备及检测方法
CN114878796B (zh) * 2022-07-12 2022-09-16 唐山陆达公路养护有限公司 基于道路养护的评估监测平台
CN115146513B (zh) * 2022-07-26 2023-04-07 北京科技大学 一种管线渗漏致塌模拟仿真预警方法和系统
CN115345892B (zh) * 2022-08-16 2025-07-11 重庆师范大学 基于相对温差的红外热图双足区域分割方法
CN115100209B (zh) * 2022-08-28 2022-11-08 电子科技大学 一种基于摄像机的图像质量的修正方法及修正系统
CN115184378B (zh) * 2022-09-15 2024-03-29 北京思莫特科技有限公司 一种基于移动设备的混凝土结构病害检测系统及方法
US12398639B2 (en) 2022-10-03 2025-08-26 Saudi Arabian Oil Company Adjusting oil and gas production based on carbon environmental impact
CN115290211B (zh) * 2022-10-10 2023-01-20 广东电网有限责任公司中山供电局 一种基于光纤传感技术的输电线环境温度测量方法
CN115898632A (zh) * 2022-11-01 2023-04-04 东南大学 基于热成像的小型柴油机实时排烟管状态检测系统
CN115762155B (zh) * 2022-11-14 2024-03-22 东南大学 一种高速公路路面异常监测方法和系统
CN115899598B (zh) * 2022-11-17 2024-06-21 浙江英集动力科技有限公司 一种融合听觉和视觉特征的供热管网状态监测方法及系统
CN115751203B (zh) * 2022-12-05 2025-06-06 南通中油燃气有限责任公司 一种基于红外热像仪的天然气管道泄漏监测系统
CN115775336A (zh) * 2022-12-06 2023-03-10 中国科学院长春光学精密机械与物理研究所 一种基于空谱特征融合的高光谱图像分类方法
CN116124836A (zh) * 2022-12-12 2023-05-16 北京城建勘测设计研究院有限责任公司 裂缝检测方法、装置、计算机设备、存储介质和程序产品
CN116046206A (zh) * 2022-12-14 2023-05-02 上海船舶电子设备研究所(中国船舶集团有限公司第七二六研究所) 环境温度自适应的分布式光纤火灾监测方法及系统
CN115931141A (zh) * 2022-12-14 2023-04-07 河南赛贝电子科技有限公司 一种基于改进型ann算法的红外测温图谱的温度识别方法
CN115823508B (zh) * 2022-12-21 2025-05-30 辽宁石油化工大学 一种空天地一体化管网安全监测方法及系统
CN115978466A (zh) * 2022-12-27 2023-04-18 重庆市荣冠科技有限公司 一种基于类别不平衡改进孪生支持向量机的流体管道泄漏检测方法
US12012082B1 (en) 2022-12-30 2024-06-18 Marathon Petroleum Company Lp Systems and methods for a hydraulic vent interlock
CN115684272B (zh) * 2023-01-03 2023-03-21 广州市市政工程试验检测有限公司 一种基于红外摄像机的钢结构裂纹检测方法
US20240218615A1 (en) * 2023-01-03 2024-07-04 Crafco, Inc. System and method for robotic sealing of defects in paved surfaces
CN116299695B (zh) * 2023-02-06 2025-07-15 海南经济特区中质猎鹰智能勘测科技有限公司 一种基于权重导向的地震属性保边滤波方法
US12006014B1 (en) 2023-02-18 2024-06-11 Marathon Petroleum Company Lp Exhaust vent hoods for marine vessels and related methods
US12043361B1 (en) 2023-02-18 2024-07-23 Marathon Petroleum Company Lp Exhaust handling systems for marine vessels and related methods
JP2024118151A (ja) * 2023-02-20 2024-08-30 横河電機株式会社 状態特定装置、状態特定方法および状態特定プログラム
CN115861825B (zh) * 2023-02-27 2023-04-25 中铁电气化局集团有限公司 一种基于图像识别的2c检测方法
CN115855070B (zh) * 2023-03-01 2023-04-28 东莞先知大数据有限公司 一种水管漏水检测方法、装置、电子设备和存储介质
CN116416270B (zh) * 2023-03-24 2024-01-23 北京城市轨道交通咨询有限公司 一种监测地下作业面渗漏水的方法及装置
CN116095914B (zh) * 2023-04-10 2023-08-25 同方德诚(山东)科技股份公司 一种基于大数据的智慧建筑照明调节方法及系统
CN116245875A (zh) * 2023-04-13 2023-06-09 南京理工大学 一种基于深度学习的地铁列车走行部温度检测方法及系统
CN116152231B (zh) * 2023-04-17 2023-07-14 卡松科技股份有限公司 基于图像处理的润滑油内杂质检测方法
US11953161B1 (en) 2023-04-18 2024-04-09 Intelcon System C.A. Monitoring and detecting pipeline leaks and spills
CN116468721A (zh) * 2023-05-11 2023-07-21 北京航空航天大学 一种基于红外的沉管隧道渗漏检测方法和装置
CN116695529A (zh) * 2023-05-17 2023-09-05 华东交通大学 一种图像处理的公路裂缝检测装置
CN116308217B (zh) * 2023-05-19 2023-08-01 中交第四航务工程勘察设计院有限公司 一种基于物联网的混凝土监控平台管理方法及系统
CN117006421A (zh) * 2023-06-02 2023-11-07 中浩建管道科技(东营)有限公司 管道泄漏监测方法、装置、设备及存储介质
CN116946645A (zh) * 2023-06-09 2023-10-27 广东科凯达智能机器人有限公司 一种基于图像识别的检测皮带撕裂情况的方法及系统
CN116403057B (zh) * 2023-06-09 2023-08-18 山东瑞盈智能科技有限公司 一种基于多源图像融合的输电线路巡检方法及系统
CN116433668B (zh) * 2023-06-14 2023-09-12 东营孚瑞特能源设备有限公司 一种智能液压油管漏油检测方法
US12297965B2 (en) 2023-08-09 2025-05-13 Marathon Petroleum Company Lp Systems and methods for mixing hydrogen with natural gas
CN116758085B (zh) * 2023-08-21 2023-11-03 山东昆仲信息科技有限公司 一种气体污染红外图像视觉辅助检测方法
CN116823839B (zh) * 2023-08-31 2023-12-01 梁山中维热力有限公司 基于热红外图像的管道泄漏检测方法
CN116862918B (zh) * 2023-09-05 2023-11-24 浙江奔月电气科技有限公司 基于人工智能的环网柜内部凝露实时检测方法
CN117169086B (zh) * 2023-09-08 2024-08-16 中建安装集团有限公司 一种建筑物地下防水层施工质量检测方法、介质及系统
US12087002B1 (en) 2023-09-18 2024-09-10 Marathon Petroleum Company Lp Systems and methods to determine depth of soil coverage along a right-of-way
US12073626B1 (en) 2023-11-22 2024-08-27 The Adt Security Corporation Analytics for detection of fluid leaks by premises monitoring systems
CN117309247B (zh) * 2023-11-28 2024-03-01 北京市农林科学院信息技术研究中心 鸡舍密闭性的判别方法、装置、系统、设备及介质
CN117392140B (zh) * 2023-12-13 2024-02-06 宏发建设有限公司 一种基于图像处理的建筑幕墙玻璃破裂检测方法及系统
CN117830300B (zh) * 2024-03-04 2024-05-14 新奥新能源工程技术有限公司 一种基于视觉的燃气管道外观质量检测方法
CN117949143B (zh) * 2024-03-26 2024-07-02 四川名人居门窗有限公司 一种门窗渗漏检测及反馈系统及方法
CN117974633B (zh) * 2024-03-28 2024-06-07 潍坊科技学院 基于图像处理的番茄病虫害智能检测方法
CN118037730B (zh) * 2024-04-12 2024-06-18 广州航海学院 一种基于计算机图像处理的裂缝长度检测系统
CN118038283B (zh) * 2024-04-15 2024-06-21 贵州黔通工程技术有限公司 一种沥青路面隐伏病害检测方法及设备
CN118668203B (zh) * 2024-05-08 2024-12-31 西南交通大学 钢轨表面疲劳裂纹与钢轨白层的一体化激光重熔修复方法
CN118154594B (zh) * 2024-05-10 2024-07-30 山东力乐包装股份有限公司 一种基于电数字数据处理的木质托盘缺陷分析系统
CN118565715B (zh) * 2024-05-27 2025-03-14 荆门美中美阀门有限公司 一种阀门密封检测方法及系统
CN118393125B (zh) * 2024-06-28 2024-09-17 欣鸿昌科技(深圳)有限公司 一种道路缺陷检测方法及检测车
CN118518713B (zh) * 2024-07-24 2024-10-11 石家庄铁道大学 隧道混凝土浇筑质量检测方法及检测终端
CN118857581B (zh) * 2024-07-31 2025-01-03 深圳市希立仪器设备有限公司 气密性检测设备的智能泄漏追踪方法及系统
CN118758503B (zh) * 2024-08-06 2025-10-03 国网安徽省电力有限公司电力科学研究院 基于单探测器的多组分绝缘气体泄漏红外检测系统及方法
CN118642548B (zh) * 2024-08-15 2024-12-20 无锡中粮工程科技有限公司 一种仓储六面主动控温方法及系统
CN119245933B (zh) * 2024-09-27 2025-09-30 上海市特种设备监督检验技术研究院 一种基于热成像技术的气体泄漏检测方法、装置及设备
CN118967684B (zh) * 2024-10-16 2025-01-10 国网甘肃省电力公司白银供电公司 基于图像识别的电力设备检测方法及系统
CN119444684B (zh) * 2024-10-21 2025-10-03 重庆大学 一种带电非接触式绝缘子劣化检测方法及装置
CN119004392B (zh) * 2024-10-23 2025-03-07 江西斯源科技股份有限公司 基于数据分析的管廊综合监控预警方法及系统
CN119713152B (zh) * 2024-10-24 2025-09-12 中国核动力研究设计院 基于高清红外的蒸汽管道泄漏检测方法及装置、存储介质
CN119027414B (zh) * 2024-10-25 2025-01-07 大连乾曜科技有限公司 一种道路施工质量检测装置及方法
CN119532643B (zh) * 2024-10-31 2025-08-08 盐城市大丰燃气设备有限公司 一种基于互联网的智慧燃气云平台
CN119068373B (zh) * 2024-11-06 2025-03-11 山东国研自动化有限公司 一种基于无人机的管道监测与维护方法
CN119516504A (zh) * 2024-11-21 2025-02-25 安徽省通途信息技术有限公司 一种基于车载边缘设备的路面病害巡检方法
CN119296067B (zh) * 2024-12-10 2025-04-29 云途信息科技(杭州)有限公司 融合多维度数据的桥头跳车检测方法及系统
CN119338802B (zh) * 2024-12-16 2025-04-01 西安航宇钛新材料科技有限公司 基于红外热力图的坯料锻造温度异常预警方法
CN119941658A (zh) * 2025-01-02 2025-05-06 新疆兵团勘测设计院集团股份有限公司 一种智慧道路在线诊断与评估系统及应用
CN119540228A (zh) * 2025-01-20 2025-02-28 浙江双久电气有限公司 一种基于图像的机械端子裂纹检测方法、介质、设备
CN120012018B (zh) * 2025-01-23 2025-08-05 中南林业科技大学 一种路面温度场参数反演和数据融合方法
CN119778569B (zh) * 2025-03-11 2025-06-13 中铁四局集团有限公司 地下管道修补方法及装置
CN119810401B (zh) * 2025-03-14 2025-06-20 北京御航智能科技有限公司 红外热像特征融合的输电线路异常区域识别方法及系统
CN119844713B (zh) * 2025-03-20 2025-08-05 浙江固特科管业有限公司 一种水暖管道维护检测方法及系统
CN119962136B (zh) * 2025-04-10 2025-07-18 中国电建集团西北勘测设计研究院有限公司 高边坡监测点动态优化布置方法
CN120043639A (zh) * 2025-04-21 2025-05-27 深圳市冠群电子有限公司 一种红外热成像机芯的低功耗散热方法及系统
CN120063929B (zh) * 2025-04-27 2025-07-22 苏州工学院 碳纤维复合三元乙丙橡胶材料受拉检测系统及其检测方法
CN120219383B (zh) * 2025-05-27 2025-07-25 西安嘉和华亨热系统有限公司 一种基于红外图像的热泵系统冷凝器工作状态监测方法
CN120253962B (zh) * 2025-06-06 2025-08-19 江西省水利科学院(江西省大坝安全管理中心、江西省水资源管理中心) 基于红外热场和瑞雷波场耦合的混凝土裂缝检测方法
CN120372295B (zh) * 2025-06-24 2025-08-19 天津大学 地下管廊温度场的评估方法及装置
CN120374610B (zh) * 2025-06-25 2025-08-22 大连艾赛机械设备有限公司 一种化工管道防泄漏安全检测方法
CN120411576A (zh) * 2025-07-02 2025-08-01 北京前往科技有限公司 基于热成像技术的公路养护设备散热性能分析系统
CN120508894B (zh) * 2025-07-15 2025-09-12 广东省特种设备检测研究院茂名检测院 一种老旧油气管道动态泄漏预测方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0961138A (ja) * 1995-08-24 1997-03-07 Mitsubishi Heavy Ind Ltd ひび割れ抽出装置
CN101701919A (zh) * 2009-11-20 2010-05-05 长安大学 一种基于图像的路面裂缝检测系统及检测方法
CN102108666A (zh) * 2011-01-17 2011-06-29 长安大学 一种沥青路面施工质量实时控制方法
CN102182137A (zh) * 2011-02-25 2011-09-14 广州飒特电力红外技术有限公司 路面缺陷检测系统及方法

Family Cites Families (65)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
SU894372A1 (ru) * 1978-06-30 1981-12-30 Центральный Ордена Трудового Красного Знамени Научно-Исследовательский Институт Черной Металлургии Им.И.П.Бардина Измеритель скорости распространени трещины в металле
JPS5629129A (en) * 1979-08-17 1981-03-23 Sharp Corp Measuring system of road surface temperature
JPS62172249A (ja) * 1986-01-25 1987-07-29 Kajima Corp 煙突の劣化診断方法及び装置
US4899296A (en) * 1987-11-13 1990-02-06 Khattak Anwar S Pavement distress survey system
US5272646A (en) * 1991-04-11 1993-12-21 Farmer Edward J Method for locating leaks in a fluid pipeline and apparatus therefore
JPH08184398A (ja) * 1995-01-05 1996-07-16 Mitsubishi Denki Bill Techno Service Kk 埋設配管漏水箇所特定方法
JP3460896B2 (ja) 1995-09-14 2003-10-27 ローランド株式会社 電子楽器の楽音生成装置
JPH1010064A (ja) * 1996-06-19 1998-01-16 Constec:Kk モルタル吹き付け法面の点検方法
AUPP107597A0 (en) * 1997-12-22 1998-01-22 Commonwealth Scientific And Industrial Research Organisation Road pavement deterioration inspection system
JP2000035372A (ja) * 1998-07-16 2000-02-02 Ishikawajima Inspection & Instrumentation Co 赤外線を用いた発泡検査方法
JP3205806B2 (ja) * 1999-04-02 2001-09-04 鹿島建設株式会社 アスファルト表面層内部の水探知方法および装置
JP2001215164A (ja) * 2000-02-02 2001-08-10 Kansai Electric Power Co Inc:The 霧状水滴を利用した真空・ガス漏れ検知装置
JP2004191258A (ja) * 2002-12-12 2004-07-08 Yoshitake Eda 建築物の漏水路検知方法
US6874932B2 (en) * 2003-06-30 2005-04-05 General Electric Company Methods for determining the depth of defects
US7073979B2 (en) * 2003-11-26 2006-07-11 Aries Industries Incorporated Method and apparatus for performing sewer maintenance with a thermal sensor
CN100402753C (zh) * 2003-12-10 2008-07-16 刘世俊 高级路面裂纹处理工艺方法
US7358860B2 (en) * 2005-03-31 2008-04-15 American Air Liquide, Inc. Method and apparatus to monitor and detect cryogenic liquefied gas leaks
CN101070947A (zh) 2006-04-28 2007-11-14 王明根 管道接缝渗漏检测系统
CN1936414A (zh) * 2006-08-12 2007-03-28 陈宜中 非导电材料供水管道检漏法
US8288726B2 (en) * 2006-12-19 2012-10-16 Weil Gary J Remote sensing of subsurface artifacts by use of visual and thermal imagery
JP5028681B2 (ja) * 2009-03-17 2012-09-19 西日本高速道路エンジニアリング四国株式会社 構造物の損傷深さ判定方法とその装置及び構造物の損傷処置判定方法とその装置
CN201449248U (zh) * 2009-03-18 2010-05-05 河海大学 一种土体裂隙发育监测仪
CN102135234A (zh) * 2010-01-27 2011-07-27 捷达世软件(深圳)有限公司 水管泄漏监控系统及方法
US20110221906A1 (en) * 2010-03-12 2011-09-15 Board Of Regents, The University Of Texas System Multiple Camera System for Automated Surface Distress Measurement
CN101845787A (zh) * 2010-04-09 2010-09-29 同济大学 基于双目视觉的水泥混凝土路面错台检测装置及方法
CN102155628A (zh) * 2010-12-01 2011-08-17 广西大学 地下排水管道渗漏检测方法及装置
US9582928B2 (en) * 2011-01-13 2017-02-28 Samsung Electronics Co., Ltd. Multi-view rendering apparatus and method using background pixel expansion and background-first patch matching
CN102374385A (zh) * 2011-07-21 2012-03-14 王斌 一种管道漏水检测装置及方法
CN102537667B (zh) * 2011-12-29 2013-09-25 杭州翰平电子技术有限公司 一种地下水管渗漏检测定位系统及方法
CN102621419A (zh) * 2012-03-28 2012-08-01 山东省电力学校 基于激光和双目视觉图像对线路电气设备自动识别和监测方法
CN102636313B (zh) * 2012-04-11 2014-12-03 浙江工业大学 基于红外热成像图像处理的渗漏源检测装置
RU2511275C2 (ru) * 2012-07-16 2014-04-10 Федеральное государственное унитарное предприятие "Научно-исследовательский институт физических проблем им. Ф.В. Лукина" Наноструктурный ик-приемник (болометр) с большой поверхностью поглощения
CN103308521A (zh) * 2012-08-29 2013-09-18 中国人民解放军第二炮兵工程大学 一种增强红外热波检测图像缺陷对比度的方法
CN102927448B (zh) * 2012-09-25 2016-12-21 北京声迅电子股份有限公司 管道无损检测方法
CN103217256A (zh) * 2013-03-20 2013-07-24 北京理工大学 基于红外图像的局部灰度-熵差的泄漏检测定位方法
CN103321129A (zh) * 2013-06-18 2013-09-25 中山市拓维电子科技有限公司 基于3g网络的红外热像的远程路面施工诊断系统及方法
CN103808760B (zh) * 2013-12-12 2017-04-26 交通运输部公路科学研究所 混凝土结构红外热成像无损检测用热激励装置
CN103882891B (zh) * 2014-01-16 2016-06-08 同济大学 利用红外热场快速预测地下连续墙侧壁渗漏的方法
CN103912791B (zh) * 2014-01-26 2016-05-04 清华大学深圳研究生院 地下管网泄漏探测方法
US9857228B2 (en) * 2014-03-25 2018-01-02 Rosemount Inc. Process conduit anomaly detection using thermal imaging
US10576907B2 (en) * 2014-05-13 2020-03-03 Gse Technologies, Llc Remote scanning and detection apparatus and method
CN104776318A (zh) * 2014-05-19 2015-07-15 白运福 一种地下水管漏水检测处理装置
CN103983513B (zh) * 2014-05-22 2016-03-02 中国矿业大学 一种采用红外辐射观测煤岩裂隙发育过程的装置及方法
CN103983514B (zh) * 2014-05-22 2016-06-01 中国矿业大学 一种煤岩裂隙发育红外辐射监测试验方法
CN104048969A (zh) * 2014-06-19 2014-09-17 樊晓东 一种隧道病害的识别方法
CN104034733A (zh) * 2014-07-02 2014-09-10 中国人民解放军国防科学技术大学 基于双目视觉监测与表面裂纹图像识别的寿命预测方法
CN104574393B (zh) * 2014-12-30 2017-08-11 北京恒达锦程图像技术有限公司 一种三维路面裂缝图像生成系统和方法
CN104713885B (zh) * 2015-03-04 2017-06-30 中国人民解放军国防科学技术大学 一种用于pcb板在线检测的结构光辅助双目测量方法
CN104749187A (zh) * 2015-03-25 2015-07-01 武汉武大卓越科技有限责任公司 基于红外温度场和灰度图像的隧道衬砌病害检测装置
CN104764528B (zh) * 2015-04-03 2018-01-12 中国矿业大学 一种煤岩裂隙发育过程中的热红外信息去噪方法
CN105113375B (zh) * 2015-05-15 2017-04-19 南京航空航天大学 一种基于线结构光的路面裂缝检测系统及其检测方法
SE539312C2 (en) * 2015-06-10 2017-06-27 Conny Andersson Med Firma Ca Konsult A method of determining the quality of a newly produced asphalt pavement
CN105465613B (zh) * 2015-11-19 2018-02-23 中建七局第二建筑有限公司 城市地下排水管道渗漏定位系统及其施工方法
CN205115977U (zh) * 2015-11-23 2016-03-30 张先 道路桥梁沥青路面裂缝检测装置
CN105719283A (zh) * 2016-01-18 2016-06-29 苏州科技学院 一种基于Hessian矩阵多尺度滤波的路面裂缝图像检测方法
CN105717163A (zh) * 2016-01-29 2016-06-29 中国商用飞机有限责任公司 红外热像检测缺陷的方法
CN105527165B (zh) * 2016-02-02 2018-07-24 山东省交通科学研究院 一种沥青路面裂缝荷载响应相对位移测试方法及测试装置
CN105938620B (zh) * 2016-04-14 2018-12-25 北京工业大学 一种小口径管内焊缝表面缺陷识别装置
CN205786366U (zh) * 2016-05-26 2016-12-07 国网浙江省电力公司宁波供电公司 一种电缆隧道水泥管片缺陷渗水特征红外热像实验装置
CN205748654U (zh) * 2016-06-21 2016-11-30 国家电网公司 基于红外热成像的变压器实时监测系统
CN106018096A (zh) * 2016-07-20 2016-10-12 中国矿业大学 煤岩破裂过程中裂隙发育区的红外辐射监测定位方法
CN106124949B (zh) * 2016-08-30 2019-08-13 国网山东省电力公司济南供电公司 一种基于热红外成像技术对绝缘子故障在线监测方法
CN206573258U (zh) 2017-02-20 2017-10-20 广东工业大学 一种管道渗漏检测装置
CN206629279U (zh) 2017-03-30 2017-11-10 中建地下空间有限公司 一种应用于综合管廊的移动巡检系统
WO2018216629A1 (ja) * 2017-05-22 2018-11-29 キヤノン株式会社 情報処理装置、情報処理方法、及びプログラム

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0961138A (ja) * 1995-08-24 1997-03-07 Mitsubishi Heavy Ind Ltd ひび割れ抽出装置
CN101701919A (zh) * 2009-11-20 2010-05-05 长安大学 一种基于图像的路面裂缝检测系统及检测方法
CN102108666A (zh) * 2011-01-17 2011-06-29 长安大学 一种沥青路面施工质量实时控制方法
CN102182137A (zh) * 2011-02-25 2011-09-14 广州飒特电力红外技术有限公司 路面缺陷检测系统及方法

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109919139A (zh) * 2019-04-01 2019-06-21 杭州晶一智能科技有限公司 基于双目立体视觉的路面状况快速检测方法
CN109903325A (zh) * 2019-04-03 2019-06-18 杭州晶一智能科技有限公司 基于立体视觉深度信息的地面精确描述方法
CN115272222A (zh) * 2022-07-25 2022-11-01 中科海微(北京)科技有限公司 道路检测信息的处理方法、装置、设备及存储介质

Also Published As

Publication number Publication date
CN109743879B (zh) 2021-02-12
WO2018122810A1 (zh) 2018-07-05
GB2581294B (en) 2021-01-06
WO2018122818A1 (zh) 2018-07-05
CN109804119B (zh) 2021-03-19
GB2573429B (en) 2020-07-15
CN109716108A (zh) 2019-05-03
GB2571016B (en) 2020-09-02
GB2581293B (en) 2021-01-06
GB201909416D0 (en) 2019-08-14
US20190339209A1 (en) 2019-11-07
GB202006107D0 (en) 2020-06-10
GB201905912D0 (en) 2019-06-12
GB2581293A (en) 2020-08-12
GB201711412D0 (en) 2017-08-30
CN109743879A (zh) 2019-05-10
CN109716108B (zh) 2021-07-13
GB2571016A (en) 2019-08-14
US20190331301A1 (en) 2019-10-31
WO2018122589A1 (zh) 2018-07-05
WO2018122809A1 (zh) 2018-07-05
CN110268190B (zh) 2021-04-30
CN109804119A (zh) 2019-05-24
CN109804232A (zh) 2019-05-24
GB2569751A (en) 2019-06-26
GB2569751B (en) 2020-04-08
GB2581294A (en) 2020-08-12
GB201905909D0 (en) 2019-06-12
CN109804232B (zh) 2021-06-01
US11221107B2 (en) 2022-01-11
CN110268190A (zh) 2019-09-20
GB202006108D0 (en) 2020-06-10
US11486548B2 (en) 2022-11-01
GB2573429A (en) 2019-11-06

Similar Documents

Publication Publication Date Title
US11486548B2 (en) System for detecting crack growth of asphalt pavement based on binocular image analysis
CN109870456B (zh) 一种路面健康状况快速检测系统及方法
CN106053475B (zh) 基于主动式全景视觉的隧道病害全断面动态快速检测装置
CN107103775B (zh) 一种基于群智计算的道路质量检测方法
KR101394244B1 (ko) 다중영상 취득장치 및 이를 활용한 이동식 노면상태 자동검지 시스템
CN110268413A (zh) 低电平传感器融合
JP6646527B2 (ja) 物体検知評価システム及び物体検知評価方法
CN111339826B (zh) 一种滑坡无人机线性传感器网络框架检测系统
CN114971166A (zh) 一种轻量化道路路面服务质量巡检系统
Chen et al. Gocomfort: Comfortable navigation for autonomous vehicles leveraging high-precision road damage crowdsensing
CN111275957A (zh) 一种交通事故信息采集方法、系统及摄像机
Rumora et al. Spatial video remote sensing for urban vegetation mapping using vegetation indices
Liu et al. Pavement surface defect recognition method based on vehicle system vibration data and feedforward neural network
Zhang et al. A two-stage correction method for UAV movement-induced errors in non-target computer vision-based displacement measurement
CN103149603A (zh) 一种基于视频的道路气象检测方法
Vogt et al. A mobile mapping solution for VRU Infrastructure monitoring via low-cost LiDAR-sensors
US12406395B2 (en) Vehicle to infrastructure extrinsic calibration system and method
Liu et al. A novel depth measurement method for urban flooding based on surveillance video images and a floating ruler
JP5407810B2 (ja) 周辺環境推定装置、周辺環境推定プログラムおよび周辺環境推定システム
Partsinevelos et al. Rockfall detection along road networks using close range photogrammetry
Tang et al. A Multi-Scene Roadside Lidar Benchmark towards Digital Twins of Road Intersections
Gong et al. DIDLM: A SLAM dataset for difficult scenarios featuring infrared, depth cameras, LiDAR, 4D radar, and others under adverse weather, low light conditions, and rough roads
Rastogi et al. Automated Construction of Time-Space Diagrams for Traffic Analysis Using Street-View Video Sequences
CN105629333A (zh) 一种基于视频的道路气象检测方法
Bassani et al. Alignment data collection of highways using mobile mapping and image analysis techniques

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17887975

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 201905912

Country of ref document: GB

Kind code of ref document: A

Free format text: PCT FILING DATE = 20171230

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17887975

Country of ref document: EP

Kind code of ref document: A1