CN116310372A - Fuzzy image recognition method, device and equipment - Google Patents

Abstract

The application discloses a fuzzy image identification method, a fuzzy image identification device and fuzzy image identification equipment. The fuzzy image recognition method comprises the steps of obtaining an image shot by shooting equipment on a wind driven generator; responding to the image comprising the blade of the wind driven generator, inputting the image into a fuzzy recognition model to obtain a fuzzy recognition result, wherein the fuzzy recognition result is used for indicating whether the image is a fuzzy image or not; the fuzzy recognition model is used for recognizing fuzzy factors in the image, wherein the fuzzy factors comprise: light-emitting elements, lens-attaching elements, and blade profile pixels. According to the embodiment of the application, whether the image comprises the blade can be firstly identified, if the image comprises the blade of the wind driven generator, the fuzzy factor in the image is identified based on the fuzzy identification model, and then whether the image is the fuzzy image is judged according to the fuzzy identification result, so that the fault condition of the blade of the wind driven generator by adopting the fuzzy image is effectively avoided, and the accuracy of detecting the blade fault of the wind driven generator is improved.

Description

Fuzzy image recognition method, device and equipment

technical field

The present application belongs to the technical field of image processing, and in particular relates to a fuzzy image recognition method, device and equipment.

Background technique

As a clean renewable energy, wind power plays an important role in the energy development strategy. Wind power mainly relies on wind generators to convert wind energy into electrical energy. When the blades of the wind turbine are running, they may encounter failures such as fracture, crack, and icing. In related technologies, it is usually detected whether these failures occur on the blades of the wind turbine by taking images of the wind turbine with a pan-tilt camera. However, when the PTZ camera is shooting, it may be affected by other factors, resulting in blurred images. If the detection is based on these blurred images, the accuracy of the detection results will often be low.

Contents of the invention

Embodiments of the present application provide a fuzzy image recognition method, device, and equipment to solve the technical problem of low accuracy in wind turbine blade fault detection.

In the first aspect, the embodiment of the present application provides a fuzzy image recognition method, the method includes:

Obtain the image captured by the capture device on the wind turbine;

In response to the image including the blade of the wind power generator, input the image into a fuzzy recognition model to obtain a fuzzy recognition result, the fuzzy recognition result is used to indicate whether the image is a fuzzy image;

Wherein, the fuzzy recognition model is used to recognize fuzzy factors in the image, and the fuzzy factors include: luminous elements, lens additional elements, and blade outline pixels.

In the second aspect, the embodiment of the present application provides a blurred image recognition device, which includes:

The obtaining module is used to obtain the image taken by the shooting device to the wind turbine;

A recognition module, configured to input the image into a fuzzy recognition model in response to the image including the blade of the wind turbine, and obtain a fuzzy recognition result, the fuzzy recognition result being used to indicate whether the image is a fuzzy image ;

Wherein, the fuzzy recognition model is used to recognize fuzzy factors in the image, and the fuzzy factors include: luminous elements, lens additional elements, and blade outline pixels.

In a third aspect, an embodiment of the present application provides an electronic device, including:

a processor and a memory storing program instructions;

The above method is implemented when the processor executes the program instructions.

In a fourth aspect, an embodiment of the present application provides a storage medium, where program instructions are stored on the storage medium, and the above method is implemented when the program instructions are executed by a processor.

In a fifth aspect, an embodiment of the present application provides a computer program product, where instructions in the computer program product are executed by a processor of an electronic device, causing the electronic device to execute the foregoing method.

The fuzzy image recognition method, device, and device of the embodiments of the present application can acquire the image taken by the shooting device of the wind power generator; in response to the blade of the wind power generator included in the image, input the image into the fuzzy recognition model to obtain the fuzzy recognition result, The fuzzy recognition result is used to indicate whether the image is a fuzzy image; wherein, the fuzzy recognition model is used to recognize the fuzzy factors in the image, and the fuzzy factors include: luminous elements, lens additional elements, and blade outline pixels. In this way, it is possible to first identify whether the image includes blades, and if the image includes blades of wind turbines, the fuzzy factors in the image are identified based on the fuzzy recognition model, and then it is judged whether the image is a fuzzy image according to the fuzzy recognition results, effectively avoiding the use of fuzzy The image detects the fault condition of the blade of the wind power generator, thereby improving the accuracy of detecting the fault of the blade of the wind power generator.

Description of drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the accompanying drawings that need to be used in the embodiments of the present application. Additional figures can be derived from these figures.

Fig. 1 is a schematic flow chart of a fuzzy image recognition method provided by an embodiment of the present application;

Fig. 2 is a schematic diagram of the working logic of the fuzzy image recognition method;

Fig. 3 is an example diagram of the network structure of the fuzzy recognition model;

Fig. 4 is a schematic diagram of fuzzy recognition model extracting semantic features;

Fig. 5 is the processing flowchart of obtaining blade area based on fuzzy recognition model;

Fig. 6 is a schematic structural diagram of a blurred image recognition device provided by another embodiment of the present application;

Fig. 7 is a schematic structural diagram of an electronic device provided by another embodiment of the present application.

Detailed ways

The characteristics and exemplary embodiments of various aspects of the application will be described in detail below. In order to make the purpose, technical solution and advantages of the application clearer, the application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described here are only intended to explain the present application rather than limit the present application. It will be apparent to one skilled in the art that the present application may be practiced without some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by showing examples of the present application.

It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is a relationship between these entities or operations. There is no such actual relationship or order between them. Furthermore, the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus comprising a set of elements includes not only those elements, but also includes elements not expressly listed. other elements of or also include elements inherent in such a process, method, article, or device. Without further limitations, an element defined by the statement "comprising..." does not exclude the presence of additional same elements in the process, method, article or device comprising said element.

In order to solve the problems in the prior art, the embodiments of the present application provide a fuzzy image recognition method, device, equipment and computer storage medium. The blurred image recognition method provided by the embodiment of the present application is firstly introduced below.

Fig. 1 shows a schematic flowchart of a blurred image recognition method provided by an embodiment of the present application. As shown in Figure 1, the fuzzy image recognition method may include the following steps:

Step 101, obtaining the image taken by the shooting device of the wind turbine;

Step 102, in response to the image including blades of wind power generators, input the image into the fuzzy recognition model to obtain a fuzzy recognition result, which is used to indicate whether the image is a fuzzy image;

Among them, the fuzzy recognition model is used to identify the fuzzy factors in the image, and the fuzzy factors include: luminous elements, lens additional elements, and blade outline pixels.

The specific implementation of the above steps will be described in detail below.

In the embodiment of the present application, the fuzzy image recognition method can obtain the image taken by the shooting device of the wind turbine; in response to the blade of the wind turbine included in the image, the image is input into the fuzzy recognition model to obtain the fuzzy recognition result, the fuzzy recognition result It is used to indicate whether the image is a blurred image; wherein, the blur identification model is used to identify blur factors in the image, and the blur factors include: luminous elements, lens additional elements, and blade outline pixels. In this way, it is possible to first identify whether the image includes blades, and if the image includes blades of wind turbines, the fuzzy factors in the image are identified based on the fuzzy recognition model, and then it is judged whether the image is a fuzzy image according to the fuzzy recognition results, effectively avoiding the use of fuzzy The image detects the fault condition of the blade of the wind power generator, thereby improving the accuracy of detecting the fault of the blade of the wind power generator.

The specific implementation of each of the above steps is introduced below.

In step 101, the image may be obtained by photographing the wind power generator through a photographing device such as a pan-tilt camera or a camera.

For example, during the working process of the wind turbine, the pan/tilt camera may record the video of the wind turbine continuously or every preset time period, and the images may be video frames extracted from the video. It may also be an image obtained by shooting the wind turbine at a preset frequency through a pan-tilt camera.

In step 102, after the image is acquired, it can be judged whether the blade of the wind turbine is included in the image. For example, the image can be input into the blade detection model obtained in advance training to obtain the blade detection result. The blade detection result can indicate the image Does it include the blades of wind turbines. The leaf detection model may specifically be a target detection model or an image classification model, which may be trained by using sample images including leaves and sample images not including leaves.

In response to the image including blades of the wind power generator, the image may be input into a fuzzy recognition model to obtain a fuzzy recognition result, and the fuzzy recognition result may be used to indicate whether the image is a fuzzy image. Specifically, the fuzzy recognition model may also be a target detection model or an image classification model, etc., and the fuzzy recognition model may be used to recognize fuzzy factors in an image.

Wherein, the fuzzy factors in the image may include luminous elements, lens additional elements and blade outline pixels. The luminous elements can refer to light sources, the sun and other luminous bodies. In other words, the fuzzy recognition model can be used to identify whether there are light sources, the sun and other luminous bodies in the image. In response to the existence of light sources, the sun and other luminous bodies in the image, the The image is a backlight image taken in a backlight environment, that is, the blur recognition result may indicate that the image is a blur image. The additional elements of the lens can be lens attachments. In other words, the fuzzy recognition model can be used to identify whether there are lens attachments such as water droplets and stains in the image. In response to the presence of lens attachments in the image, it can be considered that the PTZ camera is shooting the When the image is taken, there are attachments such as water droplets or stains on the lens, that is, the lens is not clear when shooting. At this time, the blur recognition result can indicate that the image is a blur image. The leaf contour pixels can be extracted according to the existing edge extraction algorithm. The fuzzy recognition model can be used to identify whether the leaf contour formed by the leaf contour pixels in the image is obvious. If the number of leaf contour pixels is small, the formed If the outline of the leaf is not obvious, it can be considered that the image may have poor resolution due to reasons such as focusing errors. At this time, the blurred recognition result can also indicate that the image is a blurred image.

For example, the image is input into the fuzzy recognition model, and the fuzzy recognition model can recognize the fuzzy factors of the luminescent elements, lens additional elements and blade outline pixels in the image. In response to the fuzzy factors in the image, the image can be considered as a fuzzy image, and the response Since there is no fuzzy factor in the image, it can be considered that the image is not a blurred image, that is, the image is a clear image, and the image can be used for subsequent blade fault detection.

In some examples, the related algorithm of the fuzzy recognition model can adopt the R-CNN algorithm based on the Region Proposal network, such as R-CNN, Fast R-CNN or Faster R-CNN, and these algorithms can be called secondary target detection ( Two-stage) algorithm, which is characterized by the need to use a heuristic method (selective search) or a CNN network (RPN) to generate a candidate frame (Region Proposal), and then perform classification and regression on the Region Proposal.

The relevant algorithm of the fuzzy recognition model can also use the one-stage target detection (one-stage) algorithm of Yolo or SSD, which only uses a CNN network to directly predict the categories and positions of different targets.

The accuracy of the detection results of the two-stage algorithm is relatively high, and the operation speed of the one-stage algorithm is relatively fast, and the requirements for computing resources are relatively low. In the embodiment of this application, since wind power generators are usually deployed in remote areas, it is often difficult to obtain computing resources through the Internet. Therefore, using one-stage algorithms such as Yolo can quickly detect whether the image is blurred.

In some feasible implementations, in the case of sufficient computing resources, the above-mentioned fuzzy recognition model can also use a two-stage algorithm to improve the accuracy of fuzzy image recognition results, thereby improving the accuracy of wind turbine blade fault detection. Accuracy.

As shown in Figure 2, Figure 2 is a schematic diagram of the working logic of the fuzzy image recognition method, wherein the working logic of the fuzzy image recognition method can be as follows:

Step 201, acquire an image, for example, the image may be obtained by photographing a wind power generator with a camera such as a pan-tilt camera.

Step 202 , after the image is acquired, detect whether the image includes leaves, if yes, execute step 203 , if not, execute step 207 .

Step 203, identifying the blur factors of the luminous elements and lens additional elements in the image.

Step 204 , detecting whether the image includes at least one of the luminescent element and the lens additional element, if yes, execute step 207 , if not, execute step 205 .

Step 205, identifying blur factors of blade outline pixels in the image.

Step 206 , judging that the relationship between the leaf area in the image and the leaf contour pixel points satisfies the preset contour condition, if yes, execute step 207 , if not, execute step 208 .

Step 207, determine that the image is a blurred image.

Step 208, determine that the image is not a blurred image.

For example, referring to FIG. 2, in some embodiments, the above step 102 may specifically perform the following steps:

Responsive to the image including blades of wind turbines, inputting the image into a fuzzy identification model to identify whether at least one of the luminous element and the lens additional element is included in the image;

Determining the image as a blurred image in response to the image including at least one of a luminous element and a lens additional element;

Responding to the fact that the image does not include both the luminous element and the lens additional element, obtain the blade area and the blade outline pixel points in the image;

The image is determined as a blurred image when the relationship between the area of the blade and the pixel points of the blade contour satisfies a preset contour condition.

In the embodiment of the present application, in response to the blades of wind power generators included in the image, after inputting the image into the fuzzy recognition model, it may first identify whether the image includes at least one of the luminous element and the lens additional element, if the image includes At least one of the luminescent element and the additional element of the lens, that is, the image can be considered to be an image taken under a backlight environment, or there are water droplets, stains and other attachments on the lens when the image is taken, and the image can be directly determined as a blurred image .

If the image does not include both the luminous element and the additional element of the lens, it can be considered that the image is not an image taken under a backlight environment, and the lens is clean when the image is taken, and there are no water droplets, stains and other attachments. The leaf area and leaf outline pixels in the image can be obtained to determine whether the image has poor resolution. Wherein, the leaf area in the image may be obtained based on a fuzzy recognition model, or may be obtained based on an algorithm for calculating the area of a target object in an existing image. The blade outline pixel points can be obtained based on the existing edge extraction algorithm.

After obtaining the leaf area and leaf contour pixel points in the image, it can be judged whether the relationship between the leaf area and the leaf contour pixel points satisfies the preset contour condition. For example, the contour constructed based on the blade contour pixels can be matched with the blade area, and if there is no match, it can be considered that the relationship between the blade area and the blade contour pixels satisfies the preset contour condition. For another example, the number of leaf contour pixels can be compared with the leaf area. If the ratio of the number of leaf contour pixels to the leaf area satisfies a certain threshold, it can be considered that the relationship between the leaf area and the leaf contour pixels satisfies the preset contour condition . It can be understood that specific preset contour conditions can be set according to actual conditions, and are not specifically limited here.

If the relationship between the leaf area and the pixel points of the leaf contour satisfies the preset contour condition, it can be considered that the image may have poor resolution due to reasons such as focusing errors, and the image can be determined to be a blurred image at this time. If the relationship between the leaf area and the pixel points of the leaf contour does not satisfy the preset contour condition, it can be considered that the resolution of the image is better, and the image can be determined as a clear image at this time.

In the embodiment of the present application, the blur factors of the luminous elements and lens additional elements in the image can be identified first. If the image includes at least one of the luminous elements and lens additional elements, the image can be directly determined as a blurred image. If the image The luminous element and the additional element of the lens are not included in the method, and then the blurring factors of the blade outline pixels are identified, which can effectively reduce the amount of calculation, thereby saving computing power resources, and improving the speed of blurred image recognition.

For example, referring to FIG. 2 , in some embodiments, after the above step 101, the blurred image recognition method may further include the following steps:

In response to the blades of the wind turbine not being included in the image, the image is determined to be a blurred image.

In the embodiment of the present application, after the image is acquired, it may be detected whether the image includes blades of the wind power generator. Under normal circumstances, since the image is obtained by taking pictures of the blades of the wind turbine based on a shooting device such as a pan-tilt camera, the image includes the blades of the wind turbine. However, if the image is taken at night, in a snowstorm or in heavy fog, there may not be more obvious leaf features in the image. Based on this, if the image does not include the blades of the wind turbine, it can be considered that the image may be taken at night, in a snowstorm, or in heavy fog. At this time, the image can be determined as a blurred image and cannot be used for subsequent Blade failure detection.

In some embodiments, the preset contour condition may be that the relationship between the blade area and the square of the number of blade contour pixel points satisfies a preset threshold.

It can be understood that for the image with wrong focus, due to the poor resolution of the image, the outline of the blade of the wind turbine is less obvious than the clear image, so the edge extraction algorithm in the image processing field is used to extract the blade outline pixels When pointing, the number of obtained blade outline pixels will be less, and even cause the outline to be interrupted.

It can also be understood that the square of the leaf contour length is proportional to the leaf area when the clarity of the image remains unchanged. Based on this, in order to simplify the fuzzy image recognition process, thereby saving computing power resources and improving the recognition rate, it can be based on The relationship between the leaf area and the square of the number of contour pixels is used to judge the blurring degree of the image. In other words, the preset contour condition may be that the relationship between the area of the blade and the square of the number of pixel points of the blade contour satisfies a preset threshold.

For example, if the ratio of the leaf area to the square of the number of leaf contour pixels is greater than or equal to a preset threshold, it can be considered that the number of extracted leaf contour pixels is small, that is, the resolution of the image is poor, and the image can be Determined as blurry image. The number of blade contour pixel points may refer to the number of contour pixel points in the blade length direction.

It can be understood that the preset threshold can be set based on empirical values in combination with actual conditions, or can be calculated based on multiple blurred image samples and multiple clear image samples. For example, the preset threshold can be 15.3, that is, if the ratio of the square of the leaf area in the image to the number of leaf contour pixels is greater than or equal to 15.3, the image can be considered as a blurred image, and if the ratio of the leaf area to the leaf contour pixels in the image If the ratio of the square of the number is less than 15.3, the image can be considered as a clear image.

In some embodiments, the above identification of whether the image includes at least one of the luminescent element and the lens additional element may specifically perform the following steps:

Input the image into the feature extraction network in the fuzzy recognition model, extract the features in the image through the feature extraction network, and obtain the edge features and color features in the image;

Through the path aggregation network in the fuzzy recognition model, the edge features and color features are combined to obtain the semantic features in the image;

Based on the semantic feature in the image, it is identified whether the image includes at least one of the lighting element and the lens additional element.

As shown in Figure 3, in this specific application example, the fuzzy recognition model can use the Yolo network, which can specifically include a feature extraction network, a path aggregation network, and an output network, wherein: the feature extraction network can be BackBone, which can be used to extract rough Features; the path aggregation network can be PANet, which can be used to extract multi-scale detailed features, and the fusion of related features; the output network can be recorded as Output, and its main function is to organize the output format.

As shown in Figure 4, after the image is input into the fuzzy recognition model, BackBone can extract the underlying features layer by layer for the image, such as edge features, color features, etc., where edge features can include contours and shapes, etc., and color features can include color . Exemplarily,

PANet can perform further feature extraction based on the edge features and color features extracted by BackBone, and obtain features such as "the color is gray or black or white", "the outline is obvious or not", and "the shape is in the shape of a larger wing". PANet can further combine these features to obtain semantic features in the image, such as "crack", "leaf", "stain", "cloud" and so on.

The semantic features in the image usually correspond to the objects included in the image, and whether at least one of the luminescent element and the lens additional element is included in the image can be identified through the semantic feature in the image. For example, if the semantic features in the image indicate that the image includes illuminants such as the sun and light sources, it can be considered that the image includes luminous elements. At this time, the image can be considered to be taken in a backlit environment, which is a blurred image and cannot be used for subsequent blade fault detection. For another example, the semantic features in the image indicate that the image includes lens attachments. At this time, it can be considered that the image includes additional elements of the lens. At this time, it can be considered that when the pan-tilt camera captures the image, there are attachments such as water droplets or stains on the lens. That is, the lens is not clear when shooting, and the captured image is a blurred image, which cannot be used for subsequent blade fault detection.

Output can directly output the objects included in the image based on the semantic features in the image, and can also directly output the recognition result of whether at least one of the luminous element and the additional element of the lens is included in the image.

In some examples, the detection of whether the image includes blades of wind turbines can also be performed based on the Yolo network. For example, BackBone can extract edge features, color features, etc. in the image. PANet can perform further feature extraction and feature combination based on the edge features and color features extracted by BackBone to obtain the semantic features in the image. If the semantic features in the image include "blades", it can be considered that the image includes blades of wind turbines.

In some embodiments, obtaining the leaf area in the image may include the following steps:

Through the path aggregation network in the fuzzy recognition model, the edge features and color features are combined to obtain the semantic features and position features corresponding to the leaves;

According to the semantic feature and position feature corresponding to the leaf, the area of the leaf in the image is obtained.

Generally speaking, in the Yolo network, there is usually an upsampling process, that is, filling the extracted feature map to the resolution of the input image. In the embodiment of the present application, after extracting the semantic features corresponding to the leaves, PANet can determine the position features of the leaves through upsampling, and perform fusion of the semantic features and position features, so as to obtain the leaf area in the image. Output can also output the leaf area in this image.

As shown in Figure 5, the feature extraction network can extract the first feature map from the image. As an example, the pixel size of an image can be 8×8, and the feature extraction network can perform convolution operations on the image to obtain the first feature map. At this time, the first feature map can identify objects in the image.

The path aggregation network performs further feature extraction and feature combination on the first feature map to obtain a 4×4 second feature map. At this time, the second feature map can identify large gray objects in the image.

The path aggregation network can further perform feature extraction on the second feature map to obtain a 2×2 third feature map. The third feature map can include semantic features corresponding to leaves, that is, it can be used to identify leaves in the image.

Then the path aggregation network can perform upsampling based on the second feature map and the third feature map to obtain a fourth feature map that identifies the approximate range of leaves. For example, the fourth feature map can indicate that there are leaves in the middle and lower regions of the image. For example, the path aggregation network can fill the third feature map of 2×2 to obtain a 4×4 feature map, and fuse it with the second feature map of 4×4 to obtain a discriminative leaf with a size of 4×4 A fourth feature map of the approximate range.

It can be considered that the fourth feature map includes the position feature of the blade, and the accuracy of the position feature at this time is relatively low.

The path aggregation network further upsamples the fourth feature map, and after obtaining a feature map with the same size as the first feature map, it is fused with the first feature map, because the pixel size of the first feature map can be 8×8, that is, with The pixel size of the original image is equal. At this time, the area ratio of the leaves in the image can be determined according to the fifth feature map obtained after the fusion of the first feature map and the feature map in Figure 4. For example, the fifth feature map may indicate that there are leaves with an area ratio of 20% in the middle and lower regions of the image.

Further, the path aggregation network can fuse the fifth feature map with the original image to obtain a sixth feature map that can identify the precise location of the leaf.

In the embodiment of the present application, the leaf area in the image may be acquired based on a fuzzy identification model. For example, the edge feature and color feature can be combined through the path aggregation network in the fuzzy recognition model to obtain the semantic feature corresponding to the leaf, and then the position feature of the leaf can be obtained through upsampling fusion, and then according to the semantic feature corresponding to the leaf and Position feature, get the leaf area in the image.

FIG. 6 shows a schematic structural diagram of a blurred image recognition device provided by another embodiment of the present application. For convenience of description, only parts related to the embodiment of the present application are shown.

Referring to FIG. 6, the blurred image recognition device 600 may include:

An acquisition module 601, configured to acquire images captured by the imaging device on the wind turbine;

The recognition module 602 is configured to input the image into the fuzzy recognition model in response to the blades of the wind power generator included in the image, and obtain a fuzzy recognition result, and the fuzzy recognition result is used to indicate whether the image is a fuzzy image;

Among them, the fuzzy recognition model is used to identify the fuzzy factors in the image, and the fuzzy factors include: luminous elements, lens additional elements, and blade outline pixels.

In some embodiments, the above identification module 602 may include

The identification unit is used to input the image into the fuzzy identification model in response to the image including the blade of the wind turbine, and identify whether the image includes at least one of the luminous element and the lens additional element;

a first determining unit, configured to determine the image as a blurred image in response to including at least one of a luminous element and a lens additional element in the image;

An acquisition unit, configured to acquire the leaf area and blade outline pixels in the image in response to the fact that the image does not include both the luminous element and the lens additional element;

The second determination unit is configured to determine the image as a blurred image when the relationship between the blade area and the blade contour pixel points satisfies a preset contour condition.

In some embodiments, the preset contour condition may be that the relationship between the blade area and the square of the number of blade contour pixel points satisfies a preset threshold.

In some embodiments, the identification unit can be specifically used for:

Input the image into the feature extraction network in the fuzzy recognition model, extract the features in the image through the feature extraction network, and obtain the edge features and color features in the image;

Through the path aggregation network in the fuzzy recognition model, the edge features and color features are combined to obtain the semantic features in the image;

Based on the semantic feature in the image, it is identified whether the image includes at least one of the lighting element and the lens additional element.

In some embodiments, the acquisition unit can be specifically used for:

Through the path aggregation network in the fuzzy recognition model, the edge features and color features are combined to obtain the semantic features and position features corresponding to the leaves;

According to the semantic feature and position feature corresponding to the leaf, the area of the leaf in the image is obtained.

In some embodiments, the blurred image recognition device 600 may also include:

A determining module for determining the image as a blurred image in response to the image not including blades of the wind turbine.

In some embodiments, the light emitting element is a light; and the lens attachment element is a lens attachment.

It should be noted that the information interaction and execution process between the above-mentioned devices/units are based on the same concept as the method embodiment of the present application, and are devices corresponding to the above-mentioned fuzzy image recognition method. All the implementation methods in the above-mentioned method embodiment are In the embodiments applicable to the device, its specific functions and technical effects brought by it can be referred to the method embodiments for details, and will not be repeated here.

Those skilled in the art can clearly understand that for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used for illustration. In practical applications, the above-mentioned functions can be assigned to different functional units, Completion of modules means that the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment may be integrated into one processing unit, or each unit may exist separately physically, or two or more units may be integrated into one unit, and the above-mentioned integrated units may adopt hardware It can also be implemented in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the protection scope of the present application. For the specific working processes of the units and modules in the above system, reference may be made to the corresponding processes in the aforementioned method embodiments, and details will not be repeated here.

FIG. 7 shows a schematic diagram of a hardware structure of an electronic device provided by another embodiment of the present application.

The device may include a processor 701 and a memory 702 storing program instructions.

When the processor 701 executes the program, the steps in any of the foregoing method embodiments are implemented.

Exemplarily, the program can be divided into one or more modules/units, and one or more modules/units are stored in the memory 702 and executed by the processor 701 to complete the application. One or more modules/units may be a series of program instruction segments capable of accomplishing specific functions, and the instruction segments are used to describe the execution process of the program in the device.

Specifically, the processor 701 may include a central processing unit (CPU), or an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), or may be configured to implement one or more integrated circuits in the embodiments of the present application.

Memory 702 may include mass storage for data or instructions. By way of example and not limitation, the memory 702 may include a hard disk drive (Hard Disk Drive, HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a Universal Serial Bus (Universal Serial Bus, USB) drive or two or more Combinations of multiple of the above. Storage 702 may include removable or non-removable (or fixed) media, where appropriate. Under appropriate circumstances, the storage 702 can be inside or outside the comprehensive gateway disaster recovery device. In a particular embodiment, memory 702 is a non-volatile solid-state memory.

Memory may include read only memory (ROM), random access memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical, or other physical/tangible memory storage devices. Thus, in general, memory includes one or more tangible (non-transitory) readable storage media (e.g., memory devices) encoded with software comprising computer-executable instructions, and when the software is executed (e.g., by one or more processor) operable to perform the operations described with reference to the method according to an aspect of the present disclosure.

The processor 701 reads and executes the program instructions stored in the memory 702 to implement any method in the foregoing embodiments.

In one example, the electronic device may further include a communication interface 703 and a bus 710 . Wherein, the processor 701, the memory 702, and the communication interface 703 are connected through the bus 710 and complete mutual communication.

The communication interface 703 is mainly used to realize the communication between various modules, devices, units and/or devices in the embodiments of the present application.

The bus 710 includes hardware, software or both, and couples the components of the online data traffic charging device to each other. By way of example and not limitation, the bus may include Accelerated Graphics Port (AGP) or other graphics bus, Enhanced Industry Standard Architecture (EISA) bus, Front Side Bus (FSB), HyperTransport (HT) interconnect, Industry Standard Architecture (ISA) Bus, Infiniband Interconnect, Low Pin Count (LPC) Bus, Memory Bus, Micro Channel Architecture (MCA) Bus, Peripheral Component Interconnect (PCI) Bus, PCI-Express (PCI-X) Bus, Serial Advanced Technology Attachment (SATA) bus, Video Electronics Standards Association Local (VLB) bus or other suitable bus or a combination of two or more of these. Bus 710 may comprise one or more buses, where appropriate. Although the embodiments of this application describe and illustrate a particular bus, this application contemplates any suitable bus or interconnect.

In addition, in combination with the methods in the foregoing embodiments, the embodiments of the present application may provide a storage medium for implementation. Program instructions are stored on the storage medium; when the program instructions are executed by the processor, any one of the methods in the foregoing embodiments is implemented.

The embodiment of the present application further provides a chip, the chip includes a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above method embodiments , and can achieve the same technical effect, in order to avoid repetition, it will not be repeated here.

It should be understood that the chips mentioned in the embodiments of the present application may also be called system-on-chip, system-on-chip, system-on-a-chip, or system-on-a-chip.

The embodiment of the present application provides a computer program product, the program product is stored in a storage medium, and the program product is executed by at least one processor to implement the various processes in the above method embodiments, and can achieve the same technical effect, for To avoid repetition, I won't go into details here.

It is to be understood that the application is not limited to the specific configurations and processes described above and shown in the figures. For conciseness, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art may make various changes, modifications and additions, or change the order of the steps after understanding the spirit of the present application.

The functional modules shown in the above structural block diagrams may be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it may be, for example, an electronic circuit, an application specific integrated circuit (ASIC), suitable firmware, a plug-in, a function card, or the like. When implemented in software, the elements of the present application are the programs or code segments employed to perform the required tasks. Programs or code segments can be stored in machine-readable media, or transmitted over transmission media or communication links by data signals carried in carrier waves. "Machine-readable medium" may include any medium that can store or transmit information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, and the like. Code segments may be downloaded via a computer network such as the Internet, an Intranet, or the like.

It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above steps, that is, the steps may be performed in the order mentioned in the embodiment, or may be different from the order in the embodiment, or several steps may be performed simultaneously.

Aspects of the present disclosure are described above with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and program products according to embodiments of the present disclosure. It will be understood that each block of the flowchart and/or block diagrams, and combinations of blocks in the flowchart and/or block diagrams, can be implemented by computer program instructions. These program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine such that execution of these instructions via the processor of the computer or other programmable data processing apparatus enables the Implementation of the functions/actions specified in one or more blocks of the flowchart and/or block diagrams. Such processors may be, but are not limited to, general purpose processors, special purpose processors, application specific processors, or field programmable logic circuits. It can also be understood that each block in the block diagrams and/or flowcharts and combinations of blocks in the block diagrams and/or flowcharts can also be realized by dedicated hardware for performing specified functions or actions, or can be implemented by dedicated hardware and combination of computer instructions.

The above is only a specific implementation of the present application, and those skilled in the art can clearly understand that for the convenience and brevity of the description, the specific working process of the above-described systems, modules and units can refer to the foregoing method embodiments The corresponding process in , will not be repeated here. It should be understood that the protection scope of the present application is not limited thereto, and any person familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the application, and these modifications or replacements should cover all Within the protection scope of this application.

Claims (11)

1. A blurred image recognition method, characterized by comprising:

acquiring an image shot by shooting equipment on a wind driven generator;

responding to the image comprising the blade of the wind driven generator, inputting the image into a fuzzy recognition model to obtain a fuzzy recognition result, wherein the fuzzy recognition result is used for indicating whether the image is a fuzzy image or not;

the fuzzy recognition model is used for recognizing fuzzy factors in the image, wherein the fuzzy factors comprise: light-emitting elements, lens-attaching elements, and blade profile pixels.

2. The method of claim 1, wherein the inputting the image into a fuzzy recognition model in response to the image including the blades of the wind turbine, the obtaining a fuzzy recognition result comprises:

inputting the image to a blur recognition model in response to the blade of the wind turbine being included in the image, recognizing whether at least one of a light emitting element and a lens addition element is included in the image;

determining the image as a blurred image in response to at least one of a light emitting element and a lens addition element being included in the image;

acquiring a blade area and the blade profile pixel point in the image in response to not including both a light emitting element and a lens attachment element in the image;

and determining the image as a blurred image under the condition that the relation between the blade area and the blade contour pixel points meets the preset contour condition.

3. The method of claim 2, wherein the preset profile condition is that the relationship of the blade area to the square of the number of blade profile pixels satisfies a preset threshold.

4. The method of claim 2, wherein the identifying whether at least one of a lighting element and a lens addition element is included in the image comprises:

inputting the image into a feature extraction network in a fuzzy recognition model, and extracting features in the image through the feature extraction network to obtain edge features and color features in the image;

combining the edge features and the color features through a path aggregation network in the fuzzy recognition model to obtain semantic features in the image;

based on semantic features in the image, it is identified whether at least one of a lighting element and a lens attachment element is included in the image.

5. The method of claim 4, wherein acquiring the blade area in the image comprises:

combining the edge features and the color features through a path aggregation network in the fuzzy recognition model to obtain semantic features and position features corresponding to the blades;

and acquiring the blade area in the image according to the semantic features and the position features corresponding to the blade.

6. The method of claim 1, wherein after the capturing of the image of the wind turbine by the capturing device, the method further comprises:

and determining the image as a blurred image in response to the image not including the blades of the wind turbine.

7. The method of claim 1, wherein the step of determining the position of the substrate comprises,

the light emitting element is a light emitter; and is also provided with

The lens additional element is a lens attachment.

8. A blurred image recognition apparatus, said apparatus comprising:

the acquisition module is used for acquiring an image shot by the shooting equipment on the wind driven generator;

the identification module is used for responding to the image, comprising the blades of the wind driven generator, and inputting the image into a fuzzy identification model to obtain a fuzzy identification result, wherein the fuzzy identification result is used for indicating whether the image is a fuzzy image or not;

the fuzzy recognition model is used for recognizing fuzzy factors in the image, wherein the fuzzy factors comprise: light-emitting elements, lens-attaching elements, and blade profile pixels.

9. An electronic device, the device comprising: a processor and a memory storing program instructions;

the processor, when executing the program instructions, implements the method of any one of claims 1-7.

10. A storage medium having stored thereon program instructions which, when executed by a processor, implement the method of any of claims 1-7.

11. A computer program product, characterized in that instructions in the computer program product, when executed by a processor of an electronic device, cause the electronic device to perform the method of any of claims 1-7.

Images