A distribution network unmanned aerial vehicle defect identification analysis method and system

Abstract

The application discloses a kind of network unmanned plane defect identification analysis method and system, belong to defect identification and unmanned plane technical field, including, the camera of unmanned plane is reformed, automatic focusing equipment, after unmanned plane identifies target area, equipment is photographed from multiple angles, forms line patrol picture library;The picture photographed is handled, forms high-quality line patrol picture library, and normal line and equipment are photographed from multiple angles, form standard equipment library, according to the specified classification identification forms defect data library;Using image recognition algorithm in Matlab, compare line patrol picture library with standard equipment library, mark difference picture, and compare with defect picture in defect data library one-to-many, carry out defect identification.The application quickly locks the fault range, reduces fault search time, effectively improves the efficiency of rapid power restoration, greatly reduces the personal safety risk in manual reconnaissance and improves work efficiency, improves work efficiency and work quality.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of defect identification and unmanned aerial vehicles, and particularly relates to a distribution network unmanned aerial vehicle defect identification analysis method and system. BACKGROUND

[0002] In recent years, with the continuous development of image processing technology and artificial intelligence algorithms, unmanned aerial vehicles have been increasingly widely applied in distribution network defect identification. However, the existing unmanned aerial vehicle inspection technology still has certain limitations.

[0003] On the one hand, the camera of the existing unmanned aerial vehicle often fails to achieve high-precision automatic focusing on the line and equipment during the inspection process, resulting in low-quality patrol line pictures, which affects the accuracy and efficiency of subsequent image processing. In addition, the existing image preprocessing method still has the problem of poor recognition effect in terms of contrast enhancement, edge detection and target area segmentation. These problems limit the application effect of unmanned aerial vehicles in distribution network defect identification.

[0004] On the other hand, the existing technology lacks comprehensive consideration of different weather, light and dust degree and other multi-factor, multi-dimensional lines in constructing the standard equipment library and defect data library. In terms of image recognition algorithm, although there have been some research results, they have not been fully applied in the field of unmanned aerial vehicle inspection, resulting in the need to improve the accuracy and efficiency of defect identification. SUMMARY

[0005] In view of the above-mentioned existing problems, the present application proposes a distribution network unmanned aerial vehicle defect identification analysis method for the technical problems of poor unmanned aerial vehicle inspection effect, low work efficiency, large amount of manual work for defect identification, long time for defect input system and high requirement for personnel professional technology in the existing distribution network. By modifying the camera of the unmanned aerial vehicle, optimizing the image processing process and adopting advanced image recognition algorithm, it is expected to significantly improve the accuracy and efficiency of distribution network defect identification.

[0006] To solve the above technical problems, a distribution network unmanned aerial vehicle defect identification analysis method is proposed, which comprises,

[0007] The camera of the unmanned aerial vehicle is modified to automatically focus on the equipment. After the unmanned aerial vehicle identifies the target area, the equipment is photographed from multiple angles to form a patrol line picture library. The photographed pictures are processed to form a high-quality patrol line picture library. The normal line and equipment are photographed from multiple angles to form a standard equipment library. The defect data library is formed by classifying and identifying according to the regulations. The image recognition algorithm in Matlab is used to compare the patrol line picture library with the standard equipment library, mark the difference pictures, and compare them with the defect pictures in the defect data library in one-to-many mode for defect identification.

[0008] As a preferred scheme of the distribution network unmanned plane defect identification analysis method, the line inspection picture library comprises: the camera of the unmanned plane is modified into a device for automatically focusing on the line, the fitting, the insulator and the lightning arrester, the camera is focused on the line and the device, the line or the device in the camera is effectively distinguished from the background part, image preprocessing is performed, and contrast enhancement is performed.

[0009] Edge detection is performed using the enhanced image, when the pixel value of the image after contrast enhancement is greater than μ local +k·σ local , the edge detection outputs 1 in binary, when the pixel value of the image after contrast enhancement does not exceed μ local +k·σ local , the edge detection outputs 0 in binary;

[0010] Segmentation of the target region and the background part is performed, the importance degree of the current region in the segmentation process is calculated, when the importance degree exceeds a first threshold value, the target region of the line or the device is considered, when the importance degree is less than the first threshold value, the background part is considered to be segmented and hidden, and the target region of the line or the device in the camera is quickly focused, and focusing is performed on the target region;

[0011] After the unmanned plane identifies the line or the device region, the line and the device are photographed from multiple angles to form a line inspection picture library of this line inspection.

[0012] As a preferred scheme of the distribution network unmanned plane defect identification analysis method, the high-quality line inspection picture library comprises: some influence factors are removed from the pictures in the line inspection picture library through edge detection, filtering operation, binarization, image scaling, fog removal and photo proportion in the environment of dark light, too strong light and low definition, and then the pictures are restored in details through linear correction, noise removal, bad point removal, interpolation, white balance and automatic exposure control to form the high-quality line inspection picture library.

[0013] Some noise exists in the processed image in a certain time period or a certain frame, which causes certain interference in image recognition, a binarized image E'(x,y) is obtained by binarizing the image:

[0014]

[0015] Wherein, E(x,y) is the original image, T is a noise reduction threshold, the binarized metal micro-motion image is subjected to noise reduction processing by using open loop closed operation of morphological filtering to remove noise points irrelevant to the moving object:

[0016]

[0017] Wherein, N is the working space of the moving object, B is a structural element, ° is a morphological open operation, and · is a morphological closed operation, is an erosion operation, and ⊕ is an expansion operation.

[0018] As a preferred scheme of the distribution network unmanned aerial vehicle defect identification analysis method, the standard equipment library includes, the normal line and equipment include no heating, discharge, flashover phenomenon, multi-angle shooting, forming a standard equipment library.

[0019] The defect data library includes, constructing a multi-factor, multi-dimensional line defect data library, the multi-factor includes different weather, different light, different dust degree, the multi-dimensional line includes cross-region, cross-year, cross-structure line aging, line breakage, line burr, lightning trace, insulator damage, and hardware corrosion defect picture forming defect data library.

[0020] As a preferred scheme of the distribution network unmanned aerial vehicle defect identification analysis method, the image recognition algorithm includes, image recognition comparison is carried out according to the thiesen polygon feature decomposition method, in the target distribution network area, a group of sensor nodes represented by s={s1, s2, …s n} is established through the nodes, the whole area is divided into n convex polygons through the thiesen polygon algorithm, so that each polygon contains only one node, represented as:

[0021] Π i ={q∈R 2 |d(q,s i )≤d(q,s j ),j∈{1,…,n},j≠i}

[0022] Wherein, i, j are variable indexes, q is an arbitrary point, s i And s j are any two adjacent nodes;

[0023] The picture equipment or line feature vector shot by the unmanned aerial vehicle line patrol includes the shape, color and texture of the equipment, and the feature vector of the picture shot by the unmanned aerial vehicle line patrol is compared with the line and equipment standard database to carry out equipment and line identification comparison.

[0024] As a preferred scheme of the power distribution network unmanned aerial vehicle defect identification analysis method, the comparison includes comparing the pictures in the picture library of each line inspection with the pictures in the standard equipment gallery through an algorithm in Matlab, performing nonlinear transformation on the extracted equipment or line feature vectors of the pictures taken by the unmanned aerial vehicle during line inspection, calculating the relative difference measure of the standard feature vectors, judging that there is a difference in the input pictures when the relative difference measure is greater than 0.85, and labeling the equipment state, position and defect type, and automatically screening the identified difference pictures into a difference picture folder.

[0025] As a preferred scheme of the power distribution network unmanned aerial vehicle defect identification analysis method, the comparison also includes setting the comparison threshold of the pictures screened out with differences and the defect pictures in the defect data gallery to 60%, and not performing comparison when the contrast of the pictures screened out with differences and the defect pictures in the defect data gallery is less than 60%.

[0026] When the contrast of the pictures screened out with differences and the defect pictures in the defect data gallery exceeds 60%, the same type of line or equipment defect pictures in the defect data gallery are called to perform one-to-many comparison, and the existence of multiple defects and multiple abnormalities of the same equipment defect is judged.

[0027] Another object of the present application is to provide a power distribution network unmanned aerial vehicle defect identification analysis system, which aims to solve the technical problems in unmanned aerial vehicle inspection, including improving the accuracy of image acquisition, ensuring the acquisition of clear inspection images through camera modification and image preprocessing, improving the efficiency and accuracy of defect identification, automatically comparing the standard equipment gallery and the defect data gallery through image recognition algorithms, quickly locating defects, processing the diversity and complexity of defect types, realizing accurate classification through the construction of a comprehensive defect data gallery, and optimizing the output and application of defect identification results, providing convenient defect information management for power grid operation and maintenance by screening, labeling and classifying difference pictures, thereby comprehensively improving the efficiency and safety of power grid operation and maintenance.

[0028] As a preferred scheme of the power distribution network unmanned aerial vehicle defect identification analysis system, it is characterized by comprising an image acquisition module, a defect comparison module and a defect identification module.

[0029] The image acquisition module is responsible for modifying the camera of the unmanned aerial vehicle, realizing automatic focusing, and taking pictures of the target area from multiple angles to generate a line inspection picture library, and preprocessing the taken pictures to form a high-quality line inspection picture library.

[0030] The defect comparison module constructs a standard equipment gallery and a defect data gallery, and compares the line inspection picture library with the standard equipment gallery using image recognition algorithms, labels the difference pictures, and then performs one-to-many comparison with the defect data gallery.

[0031] The defect identification module filters the identified difference pictures, labels the equipment state, position and defect type, and outputs the result to a difference picture folder.

[0032] A computer device comprises a memory and a processor, and the memory stores a computer program, characterized in that the processor implements the steps of the power distribution unmanned aerial vehicle defect identification analysis method when executing the computer program.

[0033] A computer readable storage medium stores a computer program, characterized in that the computer program is executed by a processor to implement the steps of the power distribution unmanned aerial vehicle defect identification analysis method.

[0034] Advantages of the unmanned aerial vehicle defect identification method in practical application: In terms of fault finding, the range of faults is quickly locked by combining the automatic switch action, the process of personnel climbing and crossing the river for finding is saved, the fault finding time is reduced, and the efficiency of rapid power restoration is effectively improved. In terms of disaster area investigation: for flood, forest fire and other disasters, the unmanned aerial vehicle is used to enter the disaster area for investigation multiple times, collect the on-site environment, line equipment operation and damage situation of the disaster area, and provide accurate disaster information for emergency repair and power restoration. The personal safety risk in manual investigation is greatly reduced, and the work efficiency is improved. The unmanned aerial vehicle is used for high-altitude aerial photography of traffic road expansion and municipal construction power relocation project, pre-site selection of substation, 10kV wiring reconstruction project and the like, the on-site environment of regional crossing road, river and adjacent intersection crossing, running equipment and the like is more clearly understood, so that the on-site investigation efficiency is improved, the problems of car running, on-site finding, going and not seeing are effectively solved, and the work efficiency and work quality are improved. In the new line and relocation line project completion and intermediate inspection, the unmanned aerial vehicle is used for aerial photography inspection of the conditions of tower erection, hardware cross arm installation, line binding and other key parts, which can replace the previous manual pole climbing inspection, reduce the pole climbing operation risk, improve the inspection efficiency and quality, and provide strong picture data support for power construction engineering design line selection and in-service engineering site investigation and construction scheme preparation, and improve the work efficiency and work quality. In the new line and relocation line project completion and intermediate inspection, the unmanned aerial vehicle is used for aerial photography inspection of the conditions of tower erection, hardware cross arm installation, line binding and other key parts, which can replace the previous manual pole climbing inspection, reduce the pole climbing operation risk, improve the inspection efficiency and quality, and provide strong picture data support for power construction engineering design line selection and in-service engineering site investigation and construction scheme preparation, and improve the work efficiency and work quality. In the new line and relocation line project completion and intermediate inspection, the unmanned aerial vehicle is used for aerial photography inspection of the conditions of tower erection, hardware cross arm installation, line binding and other key parts, which can replace the previous manual pole climbing inspection, reduce the pole climbing operation risk, improve the inspection efficiency and quality, and provide strong picture data support for power construction engineering design line selection and in-service engineering site investigation and construction scheme preparation, and improve the work efficiency and work quality. BRIEF DESCRIPTION OF DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0036] Figure 1A general flowchart of a distribution network unmanned aerial vehicle defect identification analysis method provided for an embodiment of the present application.

[0037] Figure 2 A Thiessen polygon algorithm diagram of a distribution network unmanned aerial vehicle defect identification analysis method provided for an embodiment of the present application.

[0038] Figure 3 A module structure schematic diagram of a distribution network unmanned aerial vehicle defect identification analysis system provided for an embodiment of the present application. DETAILED DESCRIPTION

[0039] In order to make the above objectives, features and advantages of the present application more apparent, specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work should belong to the protection scope of the present application.

[0040] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application. However, the present application can be practiced in other ways different from those described herein without departing from the scope of the present application. Those skilled in the art can make similar generalizations without departing from the scope of the present application, and therefore the present application is not limited to the specific embodiments disclosed below.

[0041] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the present application. "In one embodiment" appearing in different places in the specification does not mean the same embodiment, nor does it mean that the embodiment is mutually exclusive or selective with other embodiments.

[0042] The present application is described in detail in conjunction with the schematic diagram. In the detailed description of the embodiments of the present application, the cross-sectional view of the device structure is locally enlarged without the general proportion for the convenience of description, and the schematic diagram is only an example, which should not limit the scope of protection of the present application. In addition, the three-dimensional spatial dimensions of length, width and depth should be included in actual manufacturing.

[0043] Meanwhile, in the description of the present application, it should be noted that the terms "up, down, in and out" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the indicated device or element must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, the terms "first, second or third" are only for the purpose of description, and cannot be understood as indicating or implying relative importance.

[0044] Unless otherwise defined, the terms "mounting, connecting, associating" in the present application should be interpreted broadly, for example: it can be fixed connection, detachable connection or integral connection; it can also be mechanical connection, electrical connection or direct connection, it can also be indirectly connected through intermediate medium, or it can be the internal communication of two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0045] Embodiment 1, reference Figures 1-2 As the first embodiment of the present application, the embodiment provides a distribution network unmanned aerial vehicle defect identification analysis method, comprising:

[0046] S1: the camera of the unmanned aerial vehicle is modified to an automatic focusing device, after the unmanned aerial vehicle identifies the target area, the device is photographed from multiple angles to form a line inspection picture library.

[0047] It should be noted that the camera of the unmanned aerial vehicle is modified to an automatic focusing device of the line, the fitting, the insulator and the lightning arrester, so that the camera is concentrated on the line and the device, and the line or the device in the camera is effectively distinguished from the background part, image preprocessing is performed, and contrast enhancement is performed:

[0048]

[0049] Wherein, E(x,y) represents the pixel value of the image after contrast enhancement, I(x,y) is the value of the pixel point in the original image, μ global is the global mean of all pixel values of the original image, σ local is the standard deviation of the original image in the detection area, a is a constant to prevent the denominator from being zero, a=0.01;

[0050] The enhanced image is used for edge detection, when the pixel value E(x,y) of the image after contrast enhancement is greater than μ local +k·σ local , the edge detection binary output B(x,y) is 1, when the pixel value E(x,y) of the image after contrast enhancement does not exceed μ local +k·σ local , the edge detection binary output B(x,y) is 0, wherein k is an adjustment parameter;

[0051] Segmentation of target area and background part:

[0052]

[0053] Wherein, A represents the space size occupied by the detected area, β represents the adjustment coefficient, and S represents the importance of the current area in the segmentation process.

[0054] When S exceeds the first threshold value, it is considered to be the target area of the line or device, when S is less than the first threshold value, it is considered to be the background part to segment and hide, and the target area of the line or device in the fast focusing camera is focused, and the target area is focused:

[0055]

[0056] Where D is the position indicating the focal length of the lens, C max is the maximum limit of the visual depth, f max represents the maximum focusing range available to the device;

[0057] After the UAV identifies the line or device area, the line and device are photographed from multiple angles to form a line inspection picture library for this line inspection.

[0058] S2: Process the photographed pictures to form a high-quality line inspection picture library, take multiple-angle pictures of normal lines and devices to form a standard device picture library, and form a defect data picture library according to the specified classification and identification.

[0059] It should be noted that by removing some influencing factors through edge detection, filtering operation, binarization, image scaling, defogging, and photo proportion of the line inspection picture library, linear correction, noise removal, bad point removal, interpolation, white balance, and automatic exposure control are performed to restore the picture details, and a high-quality line inspection picture library is formed;

[0060] The processed image has some noise at a certain time period or frame, which will cause some interference during image recognition. By performing binarization processing on the image, a binarized image E'(x,y) is obtained:

[0061]

[0062] Where E(x,y) is the original image, T is the noise reduction threshold, and the open loop closed operation of morphological filtering is used to reduce the noise of the metal micro-motion image processed by binarization, and remove the noise unrelated to the moving object:

[0063]

[0064] Where N is the workspace of the moving object, B is the structure element, ° is the morphological open operation, · is the morphological closed operation, is the erosion operation, and is the dilation operation.

[0065] The specific code is as follows:

[0066]

[0067] The line inspection picture is subjected to edge detection, filtering operation, binarization, image scaling, fog removal, partial influence factor removal in the form of photo proportion, and then subjected to linear correction, noise removal, bad point removal, interpolation, white balance, automatic exposure control and other measures to restore picture details, greatly improving the quality of the picture.

[0068] It can be understood that the standard equipment gallery includes normal lines and equipment without heating, discharge and flashover phenomena, and multi-angle shooting is performed to form a standard equipment gallery.

[0069] The defect data gallery includes a multi-factor and multi-dimensional line defect data gallery, the multi-factor includes different weather, different light, and different dust levels, and the multi-dimensional line includes line aging, line breakage, line burr, lightning trace, insulator damage, and hardware corrosion defect pictures of cross-regional, cross-year, and cross-structure lines.

[0070] S3: using image recognition algorithm in Matlab, comparing the line inspection picture library with the standard equipment gallery, marking the difference pictures, and comparing with the defect pictures in the defect data gallery in one-to-many mode for defect recognition.

[0071] It should be noted that, as shown in Figure 2 , according to the image recognition comparison of the Thiessen polygon feature decomposition method, in the target distribution network area, a group of sensor nodes represented by s={s1, s2,…s n} are used, each node corresponds to a possible defect device or line component, an effective monitoring network is established through the nodes, and the entire area is divided into n convex polygons through the Thiessen polygon algorithm, so that each polygon contains only one node, represented as:

[0072] Π i ={q∈R 2 |d(q,s i )≤d(q,s j ),j∈{1,…,n},j≠i}

[0073] Wherein, i and j are variable indexes, q is an arbitrary point, s i and s j are any two adjacent nodes.

[0074] The equipment or line feature vector extracted from the picture taken by the unmanned aerial vehicle line inspection includes the shape, color and texture of the equipment, and the feature vector of the picture taken by the unmanned aerial vehicle line inspection is compared with the line and equipment standard database for equipment and line identification comparison;

[0075] Specifically, the picture in the picture library of each patrol line is compared with the picture in the standard equipment library by an algorithm in Matlab, and the extracted equipment or line feature vector of the picture taken by the unmanned aerial vehicle is nonlinearly transformed to obtain F transformed , and the relative difference measure of the standard feature vector F standard is calculated:

[0076]

[0077] When Sim relative is greater than 0.85, it is judged that the input picture has differences, and the device state, position and defect type are marked, and the recognized difference picture is automatically screened into a difference picture folder.

[0078] Further, the comparison threshold of the screened picture with the defect picture in the defect data library is set to 60%, and when the comparison degree of the screened picture with the defect picture in the defect data library is less than 60%, no comparison is performed.

[0079] When the comparison degree of the screened picture with the defect picture in the defect data library exceeds 60%, the same type of line or equipment defect picture in the defect data library is called to perform one-to-many comparison, to judge the existence of multiple defects and multiple abnormalities of the same equipment defect, greatly improving the defect recognition time and the correctness of defect recognition.

[0080] Embodiment 2, the second embodiment of the present application, which is different from the previous embodiment is:

[0081] If the function is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or say the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in the embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various storage program codes.

[0082] The logic and / or steps represented in flow diagrams or otherwise described herein, for example, can be considered as a sequence of executable instructions, and can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions, or a combination of the above. For the purposes of this specification, a "computer-readable medium" can be any apparatus that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. The computer-readable medium can be, for example but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection having one or more wires (electrical connections), a portable computer diskette (magnetic), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium can even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, for example by optically scanning the paper or other suitable medium, then electronically converted into a form that is suitable for use by the instruction execution system, apparatus, or device. The computer-readable medium can be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, for example by optically scanning the paper or other suitable medium, then electronically converted into a form that is suitable for use by the instruction execution system, apparatus, or device.

[0083] More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection having one or more wires (electrical connections), a portable computer diskette (magnetic), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium can even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, for example by optically scanning the paper or other suitable medium, then electronically converted into a form that is suitable for use by the instruction execution system, apparatus, or device.

[0084] It should be understood that aspects of the application can be implemented in hardware, software, firmware or combinations thereof. In the above embodiments, various steps or methods can be implemented, for example, using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any of the following technologies known in the art, or their combinations, can be used: discrete logic circuitry having logic gates for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc.

[0085] Embodiment 3, with reference to Figure 3 For a third embodiment of the present application, the embodiment provides a distribution network unmanned aerial vehicle defect identification analysis system, comprising an image acquisition module 10, a defect comparison module 20, a defect identification module 30;

[0086] The image acquisition module 10 is responsible for the transformation of the unmanned aerial vehicle camera, realizes automatic focusing, and multi-angle shooting of the target area, generates a line patrol picture library, and pre-processes the pictures taken to form a high-quality line patrol picture library;

[0087] The defect contrast module 20 constructs a standard equipment gallery and a defect data gallery, and contrasts the patrol picture gallery with the standard equipment gallery by using an image recognition algorithm, marks the difference pictures, and then performs one-to-many comparison with the defect data gallery;

[0088] The defect recognition module 30 screens the recognized difference pictures, marks the equipment state, position and defect type, and outputs the result to a difference picture folder.

[0089] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and all of them should be covered in the scope of the claims of the present application.

Claims

1. A method for identifying and analyzing defects in power distribution network drones, characterized in that: include, The drone's camera is modified to automatically focus on the device. After the drone identifies the target area, it takes pictures of the device from multiple angles to form a line inspection image library. The captured images are processed to create a high-quality line inspection image library. Normal lines and equipment are photographed from multiple angles to create a standard equipment image library. Defect data images are classified and identified according to regulations to create a defect data image library. Using image recognition algorithms in Matlab, the inspection image library is compared with the standard equipment image library, the difference images are marked, and a one-to-many comparison is performed with the defect images in the defect data image library to identify defects. The line inspection image library includes equipment that modifies the camera of a drone to automatically focus on lines, fittings, insulators, and surge arresters, so that the camera is focused on the lines and equipment, effectively distinguishing the lines or equipment in the camera from the background, performing image preprocessing, and enhancing the contrast. Edge detection is performed using the enhanced image; when the pixel value of the contrast-enhanced image is greater than... When edge detection is performed, the binary output is 1. When the pixel value of the image after contrast enhancement does not exceed [a certain threshold], [the result is as follows]. When an edge is detected, the binary output is 0; The target area and background are segmented, and the importance of the current area in the segmentation process is calculated. When the importance exceeds the first threshold, it is considered to be the target area of ​​the line or device. When the importance is less than the first threshold, it is considered to be the background and is segmented and hidden. The camera quickly focuses on the target area of ​​the line or device and focuses on the target area. After the drone identifies the line or equipment area, it takes pictures of the line and equipment from multiple angles to form a line inspection image library. The high-quality line patrol image library includes the following steps: removing some influencing factors by edge detection, filtering, binarization, image scaling, dehazing, and photo weighting of the images in low-light, high-light, and low-clarity environments; then performing linear correction, noise removal, bad pixel removal, interpolation, white balance, and automatic exposure control to restore image details, thus forming a high-quality line patrol image library. The image being processed contains some noise within a certain time period or frame, which can cause interference during image recognition. Binarization is used to obtain a binary image. : ； in, The image is the original image, and T is the noise reduction threshold. Morphological filtering with open-loop and closed-loop operations is used to denoise the binarized metal micro-motion image, removing noise unrelated to the moving object. ； Where N is the workspace of the moving object, B is the structuring element, ∘ is the morphological opening operation, · is the morphological closing operation, ⊗ is the erosion operation, and ⊕ is the dilation operation; The image recognition algorithm includes performing image recognition and comparison based on the Thiessen polygonal feature decomposition method, and within the target distribution network area, by... = This represents a set of sensor nodes, each corresponding to a potentially defective device or circuit component. An effective monitoring network is established through these nodes. The entire area is divided into n convex polygons using the Thiessen polygon algorithm, ensuring each polygon contains only one node. This is represented as: ； Where i and j are variable indices, Let be any point, and For any two adjacent nodes; The feature vectors of equipment or lines in the images taken by drones during line inspections are extracted, including the shape, color, and texture of the equipment. The feature vectors of the images taken by drones during line inspections are then compared with a standard database of lines and equipment to identify the equipment and lines.

2. The method for defect identification and analysis of power distribution network drones as described in claim 1, characterized in that: The standard equipment image library includes images of normal lines and equipment, including those without overheating, discharge, or flashover phenomena, taken from multiple angles to form the standard equipment image library. The defect data library includes a multi-factor, multi-dimensional defect data library for railway lines. The multi-factor factors include different weather conditions, different lighting conditions, and different degrees of dust accumulation. The multi-dimensional railway lines include images of defects such as railway aging, railway breaks, railway burrs, lightning strike marks, insulator damage, and hardware corrosion that cross regions, years, and structures.

3. The method for defect identification and analysis of power distribution network drones as described in claim 2, characterized in that: The comparison includes comparing images from the inspection image library with images from the standard equipment image library using an algorithm in Matlab, performing a nonlinear transformation on the feature vectors of the equipment or lines captured by the UAV inspection images, calculating the relative difference measure with the standard feature vectors, and when the relative difference measure is greater than 0.85, determining the differences in the input images, and labeling the equipment status, location and defect type, and automatically filtering the identified difference images into the difference image folder.

4. The method for defect identification and analysis of power distribution network drones as described in claim 3, characterized in that: The comparison also includes setting the comparison threshold between the selected images with differences and the defect images in the defect data library to 60%. When the contrast between the selected images with differences and the defect images in the defect data library is less than 60%, no comparison is performed. When the contrast between the selected image and the defect image in the defect data library exceeds 60%, images of the same type of line or equipment defect in the defect data library are retrieved for a one-to-many comparison to determine the presence of multiple defects or anomalies for the same equipment.

5. A system employing the defect identification and analysis method for power distribution network drones as described in any one of claims 1 to 4, characterized in that: It includes an image acquisition module, a defect comparison module, and a defect recognition module; The image acquisition module is responsible for modifying the drone camera to achieve automatic focusing, taking pictures of the target area from multiple angles, generating a line inspection image library, and preprocessing the pictures to form a high-quality line inspection image library. The defect comparison module constructs a standard equipment image library and a defect data image library, and uses image recognition algorithms to compare the line inspection image library with the standard equipment image library, annotates the difference images, and then performs a one-to-many comparison with the defect data image library; The defect identification module filters the identified difference images, labels the device status, location and defect type, and outputs the results to the difference image folder.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method for identifying and analyzing defects in a power distribution network drone as described in any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for identifying and analyzing defects in a power distribution network drone as described in any one of claims 1 to 4.

Images