CN107331097A - The periphery intrusion preventing apparatus and method merged based on target position information - Google Patents
The periphery intrusion preventing apparatus and method merged based on target position information Download PDFInfo
- Publication number
- CN107331097A CN107331097A CN201710647172.1A CN201710647172A CN107331097A CN 107331097 A CN107331097 A CN 107331097A CN 201710647172 A CN201710647172 A CN 201710647172A CN 107331097 A CN107331097 A CN 107331097A
- Authority
- CN
- China
- Prior art keywords
- intrusion
- target
- circumference
- periphery
- intrusion alarm
- 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.)
- Withdrawn
Links
- 230000003405 preventing effect Effects 0.000 title claims abstract description 91
- 238000000034 method Methods 0.000 title claims description 13
- 238000012544 monitoring process Methods 0.000 claims abstract description 19
- 238000004891 communication Methods 0.000 claims description 48
- 238000001514 detection method Methods 0.000 claims description 25
- 238000004422 calculation algorithm Methods 0.000 claims description 17
- 230000003287 optical effect Effects 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 9
- 230000001360 synchronised effect Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 7
- 238000007635 classification algorithm Methods 0.000 claims description 6
- 239000000835 fiber Substances 0.000 claims description 6
- 230000000007 visual effect Effects 0.000 claims description 6
- 230000036544 posture Effects 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims description 3
- 238000012423 maintenance Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 description 8
- 230000009545 invasion Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010224 classification analysis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 230000033001 locomotion Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000000149 penetrating effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000035807 sensation Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000008685 targeting Effects 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
- G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
- G08B13/19617—Surveillance camera constructional details
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
- G08B13/19654—Details concerning communication with a camera
- G08B13/19656—Network used to communicate with a camera, e.g. WAN, LAN, Internet
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
- G08B13/19663—Surveillance related processing done local to the camera
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
- G08B13/19665—Details related to the storage of video surveillance data
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The present invention proposes the periphery intrusion preventing device merged based on target position information, periphery intrusion preventing equipment is communicated with circumference intrusion alarm server, periphery intrusion preventing equipment is by circumference intrusion alarm unit, main control unit is constituted with CCTV camera, circumference intrusion alarm unit detects and positions intrusion target, warning message is sent to main control unit, main control unit is according to intrusion target quantity and positional information in warning message, dynamic sets CCTV camera focal length, determine target classification region, and judge target type, warning message is uploaded simultaneously gives circumference intrusion alarm server, circumference intrusion alarm server is according to the timestamp for receiving warning message, the video monitoring data of storage correspondence period；The target position information that the present apparatus is exported based on circumference intrusion alarm unit, dynamic adjusts CCTV camera focal length, determines video object specification area, adds object space resolution ratio, improve target classification accuracy rate, reduces the rate of false alarm of periphery intrusion preventing.
Description
Technical field
The present invention relates to a kind of periphery intrusion preventing apparatus and method, especially a kind of week merged based on target position information
Boundary's anti-intrusion device and method, belong to intelligent security guard technical field.
Background technology
Periphery intrusion preventing is protection important area, prevents the non-permitted targets such as personnel, vehicle to enter the important skill in the region
Art means.By years development, occur in that infrared emission, microwave to penetrating, tension fence, vibration optical fiber, intelligent video analysis,
（It is buried）A variety of periphery intrusion preventing technologies such as leaky cable, discrete MEMS vibrating sensors, CW with frequency modulation FMCW radars.
When rate of false alarm refers to that no target occurs, due to external interference, to cause periphery intrusion preventing system to be mistakenly considered target illegal
Invasion produces alarm, is a core technology parameter for weighing periphery intrusion preventing system practical application effect.Due to outdoor environment
Complicated, weather condition is changeable, uncontrollable disturbing factor is more, and existing periphery intrusion preventing system all suffers from how reducing system
The technological challenge of rate of false alarm.
With the development of periphery intrusion preventing technology, new periphery intrusion preventing system such as ultra-wideband leakage cable, discrete
MEMS vibrating sensors, FMCW radars etc. are respectively provided with target positioning and multiple target detection ability, and target location accuracy is general
Within 10 meters.But it is due to be disturbed by various factors under complex environment, however it remains higher rate of false alarm.
On the other hand, the intelligent video analysis technology for being conventionally used to periphery intrusion preventing is based primarily upon moving-target and background separation
Technology and artificial setting virtual perimeter domain mode realize target acquisition, and target classification ability is low, easily by illumination condition, target
The factor influence such as posture and the movement of non-targets of interest.
Develop with the technology of artificial intelligence, can have in automatic identification frame out which object and they be located at respectively
Picture target detection where turns into the study hotspot of computer vision field, successively occurs in that R-CNN, Faster R-
The many algorithms such as CNN, R-FCN, YOLO, SSD.But the accuracy rate of relative image classification up to more than 96%, because target detection is related to
And multiple links such as the search of many candidate regions, feature extraction and territorial classification, algorithm complex is high, and poor real, detection is accurate
Rate is relatively low, and the Average Accuracy for being currently based on VOC standard testing collection reaches as high as 88%.In addition, the detection of small size target is industry
The generally acknowledged problem in boundary, the Average Accuracy of the test set COCO comprising many Small objects is less than 40% at present.
The content of the invention
The problem of present invention is for existing periphery intrusion preventing rate of false alarm height influence practical application effect, is positioned with target
The circumference intrusion alarm technology of ability（Such as FMCW radars）On the basis of, combine sensation target detection technique, entered based on circumference
The target position information of warning device output, the focal length of dynamic adjustment video camera are invaded, and determines that video object classification analysis is handled
Candidate region, while increase object space resolution ratio, improve the accuracy rate of target classification, filtering out non-targeted object causes
False alarm, reduce the rate of false alarm of periphery intrusion preventing.
To realize above technical purpose, the technical scheme is that：Based on target position information merge circumference prevent into
Invade device, including circumference intrusion alarm server and the periphery intrusion preventing equipment for being deployed in zone of protection, it is characterised in that：It is described
Periphery intrusion preventing equipment is communicated by communication link with circumference intrusion alarm server, and the periphery intrusion preventing equipment includes
Circumference intrusion alarm unit, main control unit, CCTV camera, communication unit and power supply unit, the main control unit respectively with week
Boundary's intrusion alarm unit and CCTV camera connection, and be connected by communication unit with circumference intrusion alarm server, the prison
Control video camera be connected by communication unit with circumference intrusion alarm server, said supply unit respectively with circumference intrusion alarm list
Member, main control unit, CCTV camera and communication unit connection, said supply unit provide DC operation for periphery intrusion preventing equipment
Power supply；
The circumference intrusion alarm unit is used for automatic Detection location intrusion target, concurrently send warning message to main control unit, institute
The mobile processor that main control unit is Built-in Image processing unit GPU is stated, the mobile processor is responsible for processing circumference invasion report
The warning message of alert unit, intrusion target quantity and positional information in warning message, dynamic sets CCTV camera
Lens focus and the visual zone for determining target classification, and embedded target sorting algorithm of the operation based on depth network judges mesh
Type is marked, while uploading warning message gives circumference intrusion alarm server, the CCTV camera is used to monitor intrusion target institute
In the band of position, simultaneously uploaded videos monitoring data gives circumference intrusion alarm server, and the circumference intrusion alarm server is according to receipts
To the timestamp of warning message, the video monitoring data of storage correspondence period.
Further, the communication unit is fiber optic communication or ethernet communication, when periphery intrusion preventing equipment and circumference enter
When invading the distance between Alarm Server more than the farthest transmission range of Ethernet, communication link uses fiber optic communication mode；Otherwise,
Communication link uses more inexpensive ethernet communication mode.
Further, the circumference intrusion alarm server be responsible for whole circumference deployment one or more circumferences prevent
Equipment, storage, the monitor video data and circumference intrusion alarm message of retrieval circumference are invaded, and are carried out with other security protection subsystems
Data interaction and linkage；
Management of the circumference intrusion alarm server to periphery intrusion preventing equipment includes：The time of multiple periphery intrusion preventing equipment
Storage and maintenance, the collection of periphery intrusion preventing equipment state and running parameter of depth network model synchronous, for target classification
Configuration.
Further, the Content of Communication bag between the periphery intrusion preventing equipment and the circumference intrusion alarm server
Include：The monitor video encoding stream of CCTV camera is uploaded to circumference intrusion alarm server, with target type and target location
Warning message etc. data reports to circumference intrusion alarm server, periphery intrusion preventing equipment on circumference intrusion alarm server
Oneself state, periphery intrusion preventing equipment is reported to download the depth network training model updated, circumference from circumference intrusion alarm server
Intrusion alarm server（1）Setting to periphery intrusion preventing equipment parameters.
Further, when circumferential length is more than the monitoring range of single periphery intrusion preventing equipment, by disposing multiple weeks
Boundary's anti-intrusion set realizes the covering of whole circumference.
Preferably, the circumference intrusion alarm unit is FMCW radars, it is not small to the detection range of personnel targets
In 100m, target location error is no more than 5m, and the destination number that can be detected simultaneously is not less than 10.
Further, the CCTV camera is low-light (level) high-definition camera, built-in red equipped with power zoom optical lens
Outer supplementary lighting module, can carry out light filling at night.
Preferably, the video resolution of the CCTV camera is not less than 1920 × 1080, frame per second not less than 30 frames/
Second, built in video compressed encoding function, video flowing that can be after output squeezing；The night of infrared supplementary lighting module under the conditions of normal meteorological
Between irradiation distance be not less than 100m, the maximum focal length of optical lens ensure that human body target at 100m in various postures
Under pixel count be not less than 50000.
Above technical purpose is realized in order to further, the present invention also proposes a kind of week merged based on target position information
Boundary's anti-intrusion method, it is characterised in that comprise the following steps：
Step 1: periphery intrusion preventing equipment is started shooting with circumference intrusion alarm server, and network service is set up by communication unit,
The periphery intrusion preventing equipment completes the time synchronized with circumference intrusion alarm server；
Step 2: according to the scope of monitor area, circumference intrusion alarm server sets the focal length of CCTV camera, it is ensured that monitoring
Picture is clear and covers whole monitor area, and the CCTV camera is monitored to the continuous uploaded videos of circumference intrusion alarm server
Data flow, circumference intrusion alarm server storage video monitoring data；
Step 3 is according to weather conditions, and circumference intrusion alarm server dynamically sets the warning level of circumference intrusion alarm unit
Value, circumference intrusion alarm unit is detected based on current alarm threshold value to the circumference intrusion behavior in surveillance area；
Step 4: when circumference intrusion alarm unit detects intrusion behavior, on the main control unit into periphery intrusion preventing equipment
Warning message of the report comprising intrusion target quantity and positional information；
Step 5: main control unit adjusts the focal length of CCTV camera according to destination number and the positional information of each target, it is ensured that
All targets are both present in monitor video picture, while ensureing that each target has higher spatial resolution；
Step 6: main control unit determines the visual zone of target classification according to destination number and the positional information of each target, transport
Embedded target sorting algorithm of the row based on depth network, judges target type；
If Step 7: target is intrusion target, when being reported by communication unit comprising target type with positional information, alarm
Between warning message；If target is non-intrusive target, this warning message is filtered out, and upload the wrong report for including time of origin
Alert message；
Step 8: circumference intrusion alarm server is received after warning message or false alarm message, based on the timestamp in message, search
The video of rope correspondence period and storage, are recorded as alarm logging or false alarm.
Further, the main control unit in the step 6 uses different insertions according to destination number and target location
Formula target classification algorithm：If target is multiple, the position based on each target determines the candidate region of target classification respectively, adopts
Target detection is carried out with Faster R-CNN algorithms；If target is 1, the position based on the target determines object classification process
Region, using the region as a width complete image, target classification is carried out using the image classification algorithms based on depth network model.
Beneficial effects of the present invention are：
1）The target position information detected based on circumference intrusion alarm unit, dynamic sets the focal length of CCTV camera, improves mesh
Target spatial resolution（Increase size of the target in picture）, it is to avoid small size target detection problem, lift target classification
Accuracy rate, reduction circumference invasion rate of false alarm；The various sizes of training set of all kinds of targets need not be set up simultaneously, training data is reduced
Collection scale；
3）The target position information detected based on circumference intrusion alarm unit, the candidate region of sets target detection, reduces region
Quantity and search arithmetic amount, reduce the complexity of algorithm of target detection, improve algorithm real-time.
Brief description of the drawings
Fig. 1 is the overall structure diagram of periphery intrusion preventing device of the present invention.
Fig. 2 is the structured flowchart of periphery intrusion preventing equipment of the present invention.
Fig. 3 is the schematic flow sheet of periphery intrusion preventing method of the present invention.
Brief description of the drawings：1- circumference intrusion alarms server, 2- periphery intrusion preventings equipment, 21- circumference intrusion alarms unit,
22- main control units, 23- CCTV cameras, 24- communication units and 25- power supply units.
Embodiment
With reference to specific drawings and examples, the invention will be further described.
With reference to the accompanying drawings shown in 1, the periphery intrusion preventing device merged based on target position information, including circumference intrusion alarm clothes
Business device 1 and the periphery intrusion preventing equipment 2 for being deployed in zone of protection, it is characterised in that：The periphery intrusion preventing equipment 2 passes through communication
Link is communicated with circumference intrusion alarm server 1, when circumferential length is more than the monitoring range of single periphery intrusion preventing equipment 2
When, the covering of whole circumference is realized by disposing multiple periphery intrusion preventing equipment 2；
The circumference intrusion alarm server 1 is responsible for one or more periphery intrusion preventing equipment 2 of whole circumference deployment, deposits
Storage, retrieval circumference monitor video data and circumference intrusion alarm message, and with other security protection subsystems carry out data interaction and
Linkage；
Management of the circumference intrusion alarm server 1 to periphery intrusion preventing equipment 2 includes：Multiple periphery intrusion preventing equipment 2
Time synchronized, for target classification depth network model storage and maintenance, the state acquisition of periphery intrusion preventing equipment 2 with work
Parameter configuration.
With reference to the accompanying drawings shown in 2, the periphery intrusion preventing equipment 2 realize to the detection of the extremely low rate of false alarm of intrusion behavior with it is upper
Report, including circumference intrusion alarm unit 21, main control unit 22, CCTV camera 23, communication unit 24 and power supply unit 25, it is described
Main control unit 22 is connected with circumference intrusion alarm unit 21 and CCTV camera 23 respectively, and is entered by communication unit 24 and circumference
Invade Alarm Server 1 to connect, the CCTV camera 23 is connected by communication unit 24 with circumference intrusion alarm server 1, institute
Power supply unit 25 is stated to be connected with circumference intrusion alarm unit 21, main control unit 22, CCTV camera 23 and communication unit 24 respectively,
Said supply unit 25 is that periphery intrusion preventing equipment 2 provides DC supply.
Optionally, the supply voltage of circumference intrusion alarm unit 21 and CCTV camera 23 is direct current 12V, main control unit 22
Supply voltage be direct current 5V, the supply voltage of communication unit 24 is direct current 9V；
The circumference intrusion alarm unit 21 is used for automatic Detection location intrusion target, concurrently send warning message to main control unit
22, the main control unit 22 is Built-in Image processing unit GPU mobile processor, and the mobile processor is responsible for handling circumference
The warning message of intrusion alarm unit 21, intrusion target quantity and positional information in warning message, dynamic sets monitoring
The lens focus of video camera 23 and the visual zone for determining target classification, and run the embedded target classification based on depth network
Algorithm judges target type, while uploading warning message to circumference intrusion alarm server 1, the CCTV camera 23 is used to supervise
Control intrusion target position region and uploaded videos monitoring data is to circumference intrusion alarm server 1, the circumference invasion report
Alert server 1 is according to the timestamp for receiving warning message, the video monitoring data of storage correspondence period.
The circumference intrusion alarm unit 21 has automatic detection and the stationkeeping ability of multiple intrusion targets, being capable of whole day
When, round-the-clock detection intrusion target, and provide the quantity and positional information of intrusion target, the circumference intrusion alarm unit 21 is
FMCW radars, are not less than 100m, target location error is no more than 5m, can detected simultaneously to the detection range of personnel targets
Destination number is not less than 10；
The communication unit 24 is fiber optic communication or ethernet communication, when periphery intrusion preventing equipment 2 and the service of circumference intrusion alarm
When the distance between device 1 exceedes Ethernet farthest transmission range, communication link uses fiber optic communication mode；Otherwise, communication link
Using more inexpensive ethernet communication mode；
Content of Communication between the periphery intrusion preventing equipment 2 and the circumference intrusion alarm server 1 includes：CCTV camera
23 monitor video encoding stream is uploaded to circumference intrusion alarm server 1, the report with the data such as target type and target location
Alert information reporting reports itself to circumference intrusion alarm server 1, periphery intrusion preventing equipment 2 to circumference intrusion alarm server 1
State, periphery intrusion preventing equipment 2 download the depth network training model updated, circumference invasion from circumference intrusion alarm server 1
Setting of the Alarm Server 1 to the running parameter of periphery intrusion preventing equipment 2.
The CCTV camera 23 is low-light (level) high-definition camera, equipped with power zoom optical lens, built-in infrared light filling
Module, can carry out light filling at night, lift night video monitoring quality；
The video resolution of the CCTV camera 23 is not less than 1920 × 1080, and frame per second is not less than 30 frames/second, built in video pressure
Contracting encoding function, video flowing that can be after output squeezing；The night irradiation distance of infrared supplementary lighting module is not under the conditions of normal meteorological
Less than 100m, the maximum focal length of optical lens ensure that pixel count of the human body target under various postures at 100m is equal
Not less than 50000.
The main control unit 22 is Built-in Image processing unit GPU mobile processor, is responsible for processing circumference intrusion alarm
The warning message of unit 21, and according to target position information in warning message, dynamic sets the optical lens of CCTV camera 23
Focal length, makes target in video image be of moderate size, it is to avoid the target classification error under Small object size, and lifting target classification is accurate
True rate, while the various sizes of training set of various targets need not be set up, reduces training dataset scale；Invaded and reported based on circumference
Target position information in the alert warning message of unit 21, determines video analysis region, reduces candidate region quantity and selection is calculated, drop
The low complexity of algorithm of target detection, boosting algorithm real-time；Run and target is carried out based on the good depth network of training in advance
Classification.
According to Fig. 3, a kind of periphery intrusion preventing method merged based on target position information, it is characterised in that including
Following steps：
Step 1: periphery intrusion preventing equipment 2 is started shooting with circumference intrusion alarm server 1, and network is set up by communication unit 24
Communication, the periphery intrusion preventing equipment 2 completes the time synchronized with circumference intrusion alarm server 1, and the effect of time synchronized is
Ensure that timestamp skew is smaller in the intrusion alarm message that each periphery intrusion preventing equipment 2 is reported；
Periphery intrusion preventing equipment 2 carries out the time with circumference intrusion alarm server 1 using NTP NTP in the present embodiment
Synchronous, the time using circumference intrusion alarm server 1, precision was up to Millisecond as reference data；
Step 2: according to the scope of monitor area, circumference intrusion alarm server 1 sets the focal length of CCTV camera 23, it is ensured that
Monitored picture is clear and covers whole monitor area, and opens infrared built in CCTV camera 23 without blind spot, during night
Supplementary lighting module is there is provided floor light, and the CCTV camera 23 monitors number to the continuous uploaded videos of circumference intrusion alarm server 1
According to stream, the storage video monitoring data of circumference intrusion alarm server 1；
Step 3 sets the alarm of circumference intrusion alarm unit 21 according to weather conditions, the dynamic of circumference intrusion alarm server 1
Threshold value, circumference intrusion alarm unit 21 is visited based on current alarm threshold value to the circumference intrusion behavior in surveillance area
Survey；
Under the conditions of Different climate, the rate of false alarm difference of circumference intrusion alarm unit 21 is larger, and for example wind is blown near circumference
Trees, the false alarm of FMCW radars may be caused, it is necessary to for DIFFERENT METEOROLOGICAL CONDITIONS, set optimal alarm threshold value,
Incidents of false alarm is filtered out as far as possible；
Step 4: when circumference intrusion alarm unit 21 detects intrusion behavior, the main control unit into periphery intrusion preventing equipment 2
22 report the warning message comprising intrusion target quantity and positional information, and the warning message, which also includes intrusion target classification, to be believed
Breath；
Step 5: main control unit 22 adjusts the focal length of CCTV camera 23 according to destination number and the positional information of each target,
Ensure that all targets are both present in monitor video picture, while ensureing that each target has higher spatial resolution（That is mesh
It is marked on effective area in monitored picture larger）；
The principle of focal length for adjusting CCTV camera 23 is：Ensureing that all targets integrally appear in monitor video picture as far as possible
On the basis of interior, the physics size based on datum target and its distance with periphery intrusion preventing equipment 2, according to optical imagery model
Calculate the focal length of the optical lens substantially required when object pixel is counted and reaches 50000；
Step 6: main control unit 22 determines the visual zone of target classification according to destination number and the positional information of each target,
The embedded target sorting algorithm based on depth network is run, target type is judged；
The main control unit 22 uses different embedded target sorting algorithms according to destination number and target location：If mesh
It is designated as multiple, the position based on each target determines the candidate region of target classification respectively, is entered using Faster R-CNN algorithms
Row target detection；If target is 1, the position based on the target determines object classification process region, regard the region as one
Width complete image, target classification is carried out using the image classification algorithms based on depth network model.
Step 7: if target is intrusion target, by communication unit 24 report comprising target type and positional information,
The warning message of time of fire alarming；If target is non-intrusive target, this warning message is filtered out, and upload comprising time of origin
False alarm message；
Step 8: circumference intrusion alarm server 1 is received after warning message or false alarm message, based on the timestamp in message,
The video of search correspondence period and storage, are recorded as alarm logging or false alarm；
Circumference intrusion alarm server 1 in the present invention it is also possible to use alarm logging inquiry and check intrusion behavior, while using report
Picture frame training in alert record updates the depth network model of target classification, if the depth network parameter for target classification becomes
Change, then the depth network model after updating is downloaded to the main control unit 22 of periphery intrusion preventing equipment 2；False alarm note can be used simultaneously
Record carries out ex-post analysis, prevents from failing to report situation, if any failing to report, can be reviewed and disposed afterwards.
The features of the present invention is：Periphery intrusion preventing device is by being deployed in one or more circumferences of important zone of protection circumference
Anti-intrusion set 2, a set of circumference intrusion alarm server 1 are constituted, and periphery intrusion preventing equipment 2 is included with targeting capability
Circumference intrusion alarm unit 21, the CCTV camera 23 of integrated infrared supplementary lighting module and power zoom optical lens, operation vision
The main control unit 22 of target classification algorithm；Base of the present invention in the information of the detection perimeter intrusion target of circumference intrusion alarm unit 21
On plinth, the focal length of dynamic adjustment CCTV camera 23 makes the target sizes in monitor video image moderate, it is to avoid Small object size
Under target classification error, improve target classification accuracy rate；Simultaneously according to mesh in the warning message of circumference intrusion alarm unit 21
Cursor position information, it may be determined that video object analyzed area, reduces candidate region quantity and selection is calculated, and reduces target detection calculation
The complexity of method, boosting algorithm real-time.
Above to the present invention and embodiments thereof be described, the description does not have restricted, shown in accompanying drawing also only
It is one of embodiments of the present invention, actual structure is not limited thereto.If all in all ordinary skill people of this area
Member is enlightened by it, without departing from the spirit of the invention, similar to the technical scheme without designing for creativeness
Frame mode and embodiment, protection scope of the present invention all should be belonged to.
Claims (10)
1. the periphery intrusion preventing device merged based on target position information, including circumference intrusion alarm server（1）Be deployed in
The periphery intrusion preventing equipment of zone of protection（2）, it is characterised in that：The periphery intrusion preventing equipment（2）Pass through communication link and week
Boundary's intrusion alarm server（1）Communicated, the periphery intrusion preventing equipment（2）Including circumference intrusion alarm unit（21）, it is main
Control unit（22）, CCTV camera（23）, communication unit（24）And power supply unit（25）, the main control unit（22）Respectively with week
Boundary's intrusion alarm unit（21）And CCTV camera（23）Connection, and pass through communication unit（24）With circumference intrusion alarm server
（1）Connection, the CCTV camera（23）Pass through communication unit（24）With circumference intrusion alarm server（1）Connection, it is described to supply
Electric unit（25）Respectively with circumference intrusion alarm unit（21）, main control unit（22）, CCTV camera（23）And communication unit
（24）Connection, said supply unit（25）For periphery intrusion preventing equipment（2）DC supply is provided；
The circumference intrusion alarm unit（21）For automatic Detection location intrusion target, warning message is concurrently sent to main control unit
（22）, the main control unit（22）For Built-in Image processing unit GPU mobile processor, the mobile processor is responsible for processing
Circumference intrusion alarm unit（21）Warning message, intrusion target quantity and positional information in warning message are dynamically set
Put CCTV camera（23）Lens focus and determine target classification visual zone, and run based on the embedded of depth network
Target classification algorithm judges target type, while uploading warning message gives circumference intrusion alarm server（1）, the monitoring camera
Machine（23）For monitoring intrusion target position region, simultaneously uploaded videos monitoring data gives circumference intrusion alarm server（1）,
The circumference intrusion alarm server（1）According to the timestamp for receiving warning message, the video monitoring number of storage correspondence period
According to.
2. the periphery intrusion preventing device according to claim 1 merged based on target position information, it is characterised in that described
Communication unit（24）For fiber optic communication or ethernet communication, when periphery intrusion preventing equipment（2）With circumference intrusion alarm server（1）
The distance between when exceeding the farthest transmission range of Ethernet, communication link uses fiber optic communication mode；Otherwise, communication link is used
More inexpensive ethernet communication mode.
3. the periphery intrusion preventing device according to claim 1 merged based on target position information, it is characterised in that described
Circumference intrusion alarm server（1）It is responsible for one or more periphery intrusion preventing equipment of whole circumference deployment（2）, storage,
The monitor video data and circumference intrusion alarm message of circumference are retrieved, and data interaction and connection are carried out with other security protection subsystems
It is dynamic；
The circumference intrusion alarm server（1）To periphery intrusion preventing equipment（2）Management include：Multiple periphery intrusion preventing equipment
（2）Time synchronized, for target classification depth network model storage and maintenance, periphery intrusion preventing equipment（2）State is adopted
Collection is configured with running parameter.
4. the periphery intrusion preventing device according to claim 1 merged based on target position information, it is characterised in that described
Periphery intrusion preventing equipment（2）With the circumference intrusion alarm server（1）Between Content of Communication include：CCTV camera（23）
Monitor video encoding stream be uploaded to circumference intrusion alarm server（1）, report with the data such as target type and target location
Information reporting is warned to circumference intrusion alarm server（1）, periphery intrusion preventing equipment（2）To circumference intrusion alarm server（1）On
Report oneself state, periphery intrusion preventing equipment（2）From circumference intrusion alarm server（1）Download the depth network training mould updated
Type, circumference intrusion alarm server（1）To periphery intrusion preventing equipment（2）The setting of running parameter.
5. the periphery intrusion preventing device according to claim 1 merged based on target position information, it is characterised in that work as week
Boundary's length is more than single periphery intrusion preventing equipment（2）Monitoring range when, by disposing multiple periphery intrusion preventing equipment（2）Realize
The covering of whole circumference.
6. the periphery intrusion preventing device according to claim 1 merged based on target position information, it is characterised in that described
Circumference intrusion alarm unit（21）For FMCW radars, 100m is not less than to the detection range of personnel targets, target positioning is missed
Difference is no more than 5m, and the destination number that can be detected simultaneously is not less than 10.
7. the periphery intrusion preventing device according to claim 1 merged based on target position information, it is characterised in that described
CCTV camera（23）For low-light (level) high-definition camera, equipped with power zoom optical lens, built-in infrared supplementary lighting module, Neng Gou
Night carries out light filling.
8. the periphery intrusion preventing device according to claim 7 merged based on target position information, it is characterised in that described
CCTV camera（23）Video resolution be not less than 1920 × 1080, frame per second is not less than 30 frames/second, built in video compressed encoding
Function, video flowing that can be after output squeezing；The night irradiation distance of infrared supplementary lighting module is not less than under the conditions of normal meteorological
100m, the maximum focal length of optical lens ensure that pixel count of the human body target under various postures at 100m be not small
In 50000.
9. a kind of periphery intrusion preventing method merged based on target position information, it is characterised in that comprise the following steps：
Step 1: periphery intrusion preventing equipment（2）With circumference intrusion alarm server（1）Start, and pass through communication unit（24）Build
Vertical network service, the periphery intrusion preventing equipment（2）Complete and circumference intrusion alarm server（1）Time synchronized；
Step 2: according to the scope of monitor area, circumference intrusion alarm server（1）CCTV camera is set（23）Focal length,
Ensure that monitored picture is clear and covers whole monitor area, the CCTV camera（23）To circumference intrusion alarm server（1）
Continuous uploaded videos monitoring data stream, circumference intrusion alarm server（1）Store video monitoring data；
Step 3 is according to weather conditions, circumference intrusion alarm server（1）Dynamic sets circumference intrusion alarm unit（21）'s
Alarm threshold value, circumference intrusion alarm unit（21）The circumference intrusion behavior in surveillance area is entered based on current alarm threshold value
Row detection；
Step 4: when circumference intrusion alarm unit（21）When detecting intrusion behavior, to periphery intrusion preventing equipment（2）In master control
Unit（22）Report the warning message comprising intrusion target quantity and positional information；
Step 5: main control unit（22）According to destination number and the positional information of each target, CCTV camera is adjusted（23）'s
Focal length, it is ensured that all targets are both present in monitor video picture, while ensureing that each target has higher spatial resolution；
Step 6: main control unit（22）According to destination number and the positional information of each target, the visual area of target classification is determined
Domain, runs the embedded target sorting algorithm based on depth network, judges target type；
If Step 7: target is intrusion target, passing through communication unit（24）Report comprising target type and positional information, report
The warning message of alert time；If target is non-intrusive target, this warning message is filtered out, and upload comprising time of origin
False alarm message；
Step 8: circumference intrusion alarm server（1）Receive after warning message or false alarm message, based on the time in message
Stamp, the video of search correspondence period and storage, are recorded as alarm logging or false alarm.
10. a kind of periphery intrusion preventing method merged based on target position information according to claim 9, its feature is existed
In the main control unit in the step 6（22）Classified according to destination number and target location using different embedded targets
Algorithm：If target is multiple, the position based on each target determines the candidate region of target classification respectively, using Faster
R-CNN algorithms carry out target detection；If target is 1, the position based on the target determines object classification process region, by this
Region carries out target classification as a width complete image using the image classification algorithms based on depth network model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710647172.1A CN107331097A (en) | 2017-08-01 | 2017-08-01 | The periphery intrusion preventing apparatus and method merged based on target position information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710647172.1A CN107331097A (en) | 2017-08-01 | 2017-08-01 | The periphery intrusion preventing apparatus and method merged based on target position information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107331097A true CN107331097A (en) | 2017-11-07 |
Family
ID=60199823
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710647172.1A Withdrawn CN107331097A (en) | 2017-08-01 | 2017-08-01 | The periphery intrusion preventing apparatus and method merged based on target position information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107331097A (en) |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108051804A (en) * | 2017-12-19 | 2018-05-18 | 山东神戎电子股份有限公司 | A kind of inter-linked controlling method of radar and optoelectronic device |
| CN108831087A (en) * | 2018-07-17 | 2018-11-16 | 中创智慧（天津）科技发展股份有限公司 | Circumference optical fiber incudes video monitoring safety management system |
| CN109214316A (en) * | 2018-08-21 | 2019-01-15 | 北京深瞐科技有限公司 | perimeter protection method and device |
| CN109583396A (en) * | 2018-12-05 | 2019-04-05 | 广东亿迅科技有限公司 | A kind of region prevention method, system and terminal based on CNN two stages human testing |
| CN109920185A (en) * | 2019-04-16 | 2019-06-21 | 中科九度(北京)空间信息技术有限责任公司 | One kind merging the mobile mesh calibration method of detection with video data based on millimetre-wave radar |
| CN110164071A (en) * | 2019-04-29 | 2019-08-23 | 西人马（厦门）科技有限公司 | A kind of security system |
| CN110445799A (en) * | 2019-08-15 | 2019-11-12 | 杭州安恒信息技术股份有限公司 | The determination method, apparatus and server in invasion stage |
| CN111145475A (en) * | 2019-12-12 | 2020-05-12 | 上海微波技术研究所（中国电子科技集团公司第五十研究所） | Intrusion alarm system, method and medium based on vibration optical fiber and deep learning |
| CN111355922A (en) * | 2018-12-24 | 2020-06-30 | 台达电子工业股份有限公司 | Camera deployment and scheduling method, monitoring system, and non-transitory computer-readable medium |
| CN112002092A (en) * | 2020-08-07 | 2020-11-27 | 太原理工大学 | High-speed railway platform anti-border-crossing distributed optical fiber early warning system |
| CN112083678A (en) * | 2020-08-25 | 2020-12-15 | 深圳市格灵人工智能与机器人研究院有限公司 | Gate security control method and device, electronic equipment and storage medium |
| CN112136132A (en) * | 2018-05-22 | 2020-12-25 | 三菱电机株式会社 | Installation location selection support device, installation location selection support method, and installation location selection support program |
| CN113033521A (en) * | 2021-05-25 | 2021-06-25 | 南京甄视智能科技有限公司 | Perimeter dynamic early warning method and system based on target analysis |
| CN113110261A (en) * | 2021-05-23 | 2021-07-13 | 杭州翔毅科技有限公司 | Man-machine interaction system integrating artificial intelligence and manufacturing industry |
| CN113534276A (en) * | 2021-07-12 | 2021-10-22 | 西安电子科技大学 | Railway intrusion detection method based on Faster R-CNN |
| CN114495398A (en) * | 2020-11-13 | 2022-05-13 | 北京机械设备研究所 | Intelligent protection device and protection method under outdoor unattended condition |
| CN114913481A (en) * | 2022-06-01 | 2022-08-16 | 无锡八英里电子科技有限公司 | Gas station/charging station abnormal intrusion detection method based on image recognition |
| CN115063940A (en) * | 2022-06-06 | 2022-09-16 | 中国银行股份有限公司 | Risk monitoring method and device, storage medium and electronic equipment |
| CN115294207A (en) * | 2022-06-30 | 2022-11-04 | 南京南邮信息产业技术研究院有限公司 | Fusion scheduling system and method of smart campus surveillance video and 3D GIS model |
| CN115662037A (en) * | 2022-10-09 | 2023-01-31 | 通号通信信息集团有限公司 | Method and system for railway perimeter protection |
| CN115909215A (en) * | 2022-12-09 | 2023-04-04 | 厦门农芯数字科技有限公司 | An edge intrusion early warning method and system based on target detection |
| CN116007684A (en) * | 2023-02-15 | 2023-04-25 | 四川锦美环保股份有限公司 | Intelligent unmanned supervision system and method for drinking water source |
| CN117333995A (en) * | 2023-09-20 | 2024-01-02 | 华能核能技术研究院有限公司 | Yolo-v 3-based high-temperature gas cooled reactor perimeter intrusion detection alarm method and system |
| CN117496648A (en) * | 2023-10-30 | 2024-02-02 | 武汉大学 | Electronic alert system, server, storage media |
| CN117830053A (en) * | 2023-12-22 | 2024-04-05 | 中国电子信息产业集团有限公司第六研究所 | Perimeter security alarm system and method |
| CN118803208A (en) * | 2024-09-12 | 2024-10-18 | 天津津轨汇海科技发展有限公司 | Linkage control system and method for camera and perimeter security based on single chip microcomputer |
| CN119296238A (en) * | 2024-10-15 | 2025-01-10 | 北京红山信息科技研究院有限公司 | A railway perimeter intrusion monitoring and warning system |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000003478A (en) * | 1998-06-16 | 2000-01-07 | Mitsubishi Electric Corp | Motion detection device |
| AU2003100055A4 (en) * | 2003-01-28 | 2003-07-03 | Ronald Stephen Abernethy | Graffiti Guard |
| JP2005057743A (en) * | 2003-07-22 | 2005-03-03 | Hitachi Kokusai Electric Inc | Object tracking method and object tracking device |
| US20070230744A1 (en) * | 2006-03-29 | 2007-10-04 | Mark Dronge | Security alarm system |
| CN106157502A (en) * | 2016-08-28 | 2016-11-23 | 烟台三环锁业集团股份有限公司 | A kind of based on multi-sensor fusion technology Initiative Defense intelligence intrusion detection equipment |
| WO2017036363A1 (en) * | 2015-09-02 | 2017-03-09 | 同方威视技术股份有限公司 | Optical fiber perimeter intrusion signal identification method and device, and perimeter intrusion alarm system |
| CN106600872A (en) * | 2017-01-10 | 2017-04-26 | 秦皇岛博微智能科技有限公司 | Radar video linkage based intelligent boundary security system |
| CN207164939U (en) * | 2017-08-01 | 2018-03-30 | 中科融通物联科技无锡有限公司 | Periphery intrusion preventing device based on target position information fusion |
-
2017
- 2017-08-01 CN CN201710647172.1A patent/CN107331097A/en not_active Withdrawn
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000003478A (en) * | 1998-06-16 | 2000-01-07 | Mitsubishi Electric Corp | Motion detection device |
| AU2003100055A4 (en) * | 2003-01-28 | 2003-07-03 | Ronald Stephen Abernethy | Graffiti Guard |
| JP2005057743A (en) * | 2003-07-22 | 2005-03-03 | Hitachi Kokusai Electric Inc | Object tracking method and object tracking device |
| US20070230744A1 (en) * | 2006-03-29 | 2007-10-04 | Mark Dronge | Security alarm system |
| WO2017036363A1 (en) * | 2015-09-02 | 2017-03-09 | 同方威视技术股份有限公司 | Optical fiber perimeter intrusion signal identification method and device, and perimeter intrusion alarm system |
| CN106157502A (en) * | 2016-08-28 | 2016-11-23 | 烟台三环锁业集团股份有限公司 | A kind of based on multi-sensor fusion technology Initiative Defense intelligence intrusion detection equipment |
| CN106600872A (en) * | 2017-01-10 | 2017-04-26 | 秦皇岛博微智能科技有限公司 | Radar video linkage based intelligent boundary security system |
| CN207164939U (en) * | 2017-08-01 | 2018-03-30 | 中科融通物联科技无锡有限公司 | Periphery intrusion preventing device based on target position information fusion |
Cited By (35)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108051804A (en) * | 2017-12-19 | 2018-05-18 | 山东神戎电子股份有限公司 | A kind of inter-linked controlling method of radar and optoelectronic device |
| CN112136132A (en) * | 2018-05-22 | 2020-12-25 | 三菱电机株式会社 | Installation location selection support device, installation location selection support method, and installation location selection support program |
| CN108831087A (en) * | 2018-07-17 | 2018-11-16 | 中创智慧（天津）科技发展股份有限公司 | Circumference optical fiber incudes video monitoring safety management system |
| CN109214316A (en) * | 2018-08-21 | 2019-01-15 | 北京深瞐科技有限公司 | perimeter protection method and device |
| CN109583396A (en) * | 2018-12-05 | 2019-04-05 | 广东亿迅科技有限公司 | A kind of region prevention method, system and terminal based on CNN two stages human testing |
| CN111355922B (en) * | 2018-12-24 | 2021-08-17 | 台达电子工业股份有限公司 | Camera deployment and scheduling method, monitoring system, and non-transitory computer-readable medium |
| CN111355922A (en) * | 2018-12-24 | 2020-06-30 | 台达电子工业股份有限公司 | Camera deployment and scheduling method, monitoring system, and non-transitory computer-readable medium |
| CN109920185A (en) * | 2019-04-16 | 2019-06-21 | 中科九度(北京)空间信息技术有限责任公司 | One kind merging the mobile mesh calibration method of detection with video data based on millimetre-wave radar |
| CN110164071A (en) * | 2019-04-29 | 2019-08-23 | 西人马（厦门）科技有限公司 | A kind of security system |
| CN110445799B (en) * | 2019-08-15 | 2021-11-05 | 杭州安恒信息技术股份有限公司 | Method, device and server for determining intrusion stage |
| CN110445799A (en) * | 2019-08-15 | 2019-11-12 | 杭州安恒信息技术股份有限公司 | The determination method, apparatus and server in invasion stage |
| CN111145475A (en) * | 2019-12-12 | 2020-05-12 | 上海微波技术研究所（中国电子科技集团公司第五十研究所） | Intrusion alarm system, method and medium based on vibration optical fiber and deep learning |
| CN112002092A (en) * | 2020-08-07 | 2020-11-27 | 太原理工大学 | High-speed railway platform anti-border-crossing distributed optical fiber early warning system |
| CN112002092B (en) * | 2020-08-07 | 2022-02-01 | 太原理工大学 | High-speed railway platform anti-border-crossing distributed optical fiber early warning system |
| CN112083678A (en) * | 2020-08-25 | 2020-12-15 | 深圳市格灵人工智能与机器人研究院有限公司 | Gate security control method and device, electronic equipment and storage medium |
| CN114495398A (en) * | 2020-11-13 | 2022-05-13 | 北京机械设备研究所 | Intelligent protection device and protection method under outdoor unattended condition |
| CN113110261A (en) * | 2021-05-23 | 2021-07-13 | 杭州翔毅科技有限公司 | Man-machine interaction system integrating artificial intelligence and manufacturing industry |
| CN113033521B (en) * | 2021-05-25 | 2022-05-10 | 南京甄视智能科技有限公司 | Perimeter dynamic early warning method and system based on target analysis |
| CN113033521A (en) * | 2021-05-25 | 2021-06-25 | 南京甄视智能科技有限公司 | Perimeter dynamic early warning method and system based on target analysis |
| CN113534276A (en) * | 2021-07-12 | 2021-10-22 | 西安电子科技大学 | Railway intrusion detection method based on Faster R-CNN |
| CN113534276B (en) * | 2021-07-12 | 2022-06-07 | 西安电子科技大学 | Railway intrusion behavior detection method based on fast R-CNN |
| CN114913481B (en) * | 2022-06-01 | 2023-05-26 | 无锡八英里电子科技有限公司 | Image recognition-based abnormal intrusion detection method for gas station or charging station |
| CN114913481A (en) * | 2022-06-01 | 2022-08-16 | 无锡八英里电子科技有限公司 | Gas station/charging station abnormal intrusion detection method based on image recognition |
| CN115063940A (en) * | 2022-06-06 | 2022-09-16 | 中国银行股份有限公司 | Risk monitoring method and device, storage medium and electronic equipment |
| CN115063940B (en) * | 2022-06-06 | 2024-02-09 | 中国银行股份有限公司 | Risk monitoring method and device, storage medium and electronic equipment |
| CN115294207A (en) * | 2022-06-30 | 2022-11-04 | 南京南邮信息产业技术研究院有限公司 | Fusion scheduling system and method of smart campus surveillance video and 3D GIS model |
| CN115662037A (en) * | 2022-10-09 | 2023-01-31 | 通号通信信息集团有限公司 | Method and system for railway perimeter protection |
| CN115909215A (en) * | 2022-12-09 | 2023-04-04 | 厦门农芯数字科技有限公司 | An edge intrusion early warning method and system based on target detection |
| CN116007684A (en) * | 2023-02-15 | 2023-04-25 | 四川锦美环保股份有限公司 | Intelligent unmanned supervision system and method for drinking water source |
| CN117333995A (en) * | 2023-09-20 | 2024-01-02 | 华能核能技术研究院有限公司 | Yolo-v 3-based high-temperature gas cooled reactor perimeter intrusion detection alarm method and system |
| CN117496648A (en) * | 2023-10-30 | 2024-02-02 | 武汉大学 | Electronic alert system, server, storage media |
| CN117830053A (en) * | 2023-12-22 | 2024-04-05 | 中国电子信息产业集团有限公司第六研究所 | Perimeter security alarm system and method |
| CN118803208A (en) * | 2024-09-12 | 2024-10-18 | 天津津轨汇海科技发展有限公司 | Linkage control system and method for camera and perimeter security based on single chip microcomputer |
| CN118803208B (en) * | 2024-09-12 | 2025-01-17 | 天津津轨汇海科技发展有限公司 | Linkage control system and method for camera and perimeter security based on single chip microcomputer |
| CN119296238A (en) * | 2024-10-15 | 2025-01-10 | 北京红山信息科技研究院有限公司 | A railway perimeter intrusion monitoring and warning system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107331097A (en) | The periphery intrusion preventing apparatus and method merged based on target position information | |
| CN207164939U (en) | Periphery intrusion preventing device based on target position information fusion | |
| CN101388130B (en) | Multi-level and three-dimensional enclosure anti-intrusion system, device and implementation method based on wireless sensor network | |
| CN103606236B (en) | A kind of Real-time security monitoring system based on distributed fiber optic intrusion alarm unit | |
| CN112288984A (en) | Three-dimensional visual unattended substation intelligent linkage system based on video fusion | |
| CN101607668B (en) | Embedded computer vision escalator pedestrian flow supervision and alarm device | |
| CN105100757B (en) | Detector with integral structure | |
| CN202947884U (en) | Motor train unit operation fault dynamic image detection system | |
| CN104613892B (en) | Merge the compound snow depth monitoring system of video detection technology and laser ranging technique | |
| US20080172781A1 (en) | System and method for obtaining and using advertising information | |
| CN104533525B (en) | Driving face in coal mine coal and gas prominent alarm method based on image | |
| KR102434154B1 (en) | Method for tracking multi target in traffic image-monitoring-system | |
| CN105763853A (en) | Emergency early warning method for stampede accident in public area | |
| KR101149916B1 (en) | Warning system of landslide | |
| CN210899299U (en) | Tunnel monitoring system | |
| CN102646309A (en) | Intelligent video perimeter rail system and control method thereof | |
| CN103929592A (en) | All-dimensional intelligent monitoring equipment and method | |
| CN114283544B (en) | An artificial intelligence-based railway platform intrusion monitoring system and method | |
| CN106454226A (en) | 360-degree panorama technology based integrated pipe corridor online monitoring and alarming method and system | |
| CN104533526B (en) | Coal-face coal and gas prominent alarm method based on image | |
| KR20220000172A (en) | An apparatus and a system for providing a security surveillance service based on edge computing and a method for operating them | |
| KR101454644B1 (en) | Loitering Detection Using a Pedestrian Tracker | |
| KR20220000226A (en) | A system for providing a security surveillance service based on edge computing | |
| CN108206931A (en) | A kind of legacy monitoring analysis system | |
| CN110868568A (en) | Video monitoring method and device for station outside air pipe |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20171107 |
|
| WW01 | Invention patent application withdrawn after publication |