A method and system for defect detection of lines in a power distribution network

Abstract

The application discloses a kind of defect detection method and system for line in power distribution network, method includes: each surface image in the same surface image set is compared, according to the comparison result in the same surface image set Target surface image is selected, target surface image sequence is constructed, and target surface image sequence is divided based on the preset interception rule, at least one target surface image subsequence is obtained, in a certain target surface image subsequence Any target surface image is extracted, and any target surface image is identified based on the preset image recognition model, and whether the first identification result of image recognition model output is abnormal is judged. In the face of the scratch situation of the continuity of the line to be detected, the identification efficiency of the line to be detected can be effectively improved.

Description

Technical Field

[0001] This invention belongs to the field of power distribution network monitoring technology, and particularly relates to a method and system for detecting defects in power distribution lines. Background Technology

[0002] Electricity is the lifeblood of economic development, providing support and security for industry, agriculture, and urban construction. The rapid development of smart grids and the ubiquitous power Internet of Things has provided a broad platform for the application of artificial intelligence technology. Data-driven power AI technology is playing an increasingly important role and has become one of the key strategic directions for power grid development.

[0003] The safe operation of power distribution lines is one of the foundations of the safe operation of the entire power distribution network. Therefore, the inspection of power distribution lines is the most important part of the power distribution network inspection work. With the continuous development of the economy and the accelerated construction of power distribution networks, there are more and more defects and faults in power distribution lines. As a result, the requirements for preventive maintenance are also constantly increasing.

[0004] In existing technologies, each surface of the line segment to be inspected needs to be identified sequentially, which increases the amount of identification required and thus reduces the efficiency of defect detection. Summary of the Invention

[0005] This invention provides a method and system for defect detection of lines in a power distribution network, which solves the technical problem of needing to identify each surface of the line segment to be inspected in sequence.

[0006] In a first aspect, the present invention provides a method for defect detection of lines in a power distribution network, comprising:

[0007] Acquire a surface image set of at least one line segment to be detected, wherein the surface image set contains surface images in at least one direction;

[0008] Compare the surface images in the same surface image set, select the target surface image in the same surface image set according to the comparison results, and construct the association relationship between the target surface image and other surface images, wherein the other surface images are the surface images in the same surface image set excluding the target surface image;

[0009] Based on the location information of the at least one line segment to be detected on a preset electronic map, sort the target surface images to obtain a target surface image sequence.

[0010] The target surface image sequence is divided according to a preset cropping rule to obtain at least one target surface image subsequence, wherein each target surface image in a target surface image subsequence belongs to the same direction;

[0011] Extract any target surface image from a subsequence of target surface images, identify the target surface image based on a preset image recognition model, and determine whether the first recognition result output by the image recognition model is abnormal.

[0012] If the first recognition result is not abnormal, then at least one first other surface image associated with any of the target surface images is obtained, a second recognition result of the at least one first other surface image is obtained according to the image recognition model, and a second other surface image associated with other target surface images in a certain target surface image subsequence is recognized according to the second recognition result to obtain a third recognition result;

[0013] Based on the second and third identification results, it is determined whether the line segment to be detected is abnormal, corresponding to each target surface image in the subsequence of a certain target surface image.

[0014] Secondly, the present invention provides a defect detection system for lines in a power distribution network, comprising:

[0015] The acquisition module is configured to acquire at least one set of surface images of a line segment to be detected, wherein the set of surface images includes surface images in at least one direction;

[0016] The construction module is configured to compare various surface images in the same surface image set, select a target surface image in the same surface image set according to the comparison results, and construct the association relationship between the target surface image and other surface images, wherein the other surface images are surface images in the same surface image set excluding the target surface image;

[0017] The sorting module is configured to sort each target surface image according to the location information of the at least one line segment to be detected on a preset electronic map, so as to obtain a target surface image sequence.

[0018] The segmentation module is configured to segment the target surface image sequence based on a preset truncation rule to obtain at least one target surface image subsequence, wherein each target surface image in a target surface image subsequence belongs to the same direction;

[0019] The first recognition module is configured to extract any target surface image from a subsequence of target surface images, recognize the any target surface image based on a preset image recognition model, and determine whether the first recognition result output by the image recognition model is abnormal.

[0020] The second recognition module is configured to, if the first recognition result is not abnormal, acquire at least one first other surface image associated with any target surface image, acquire a second recognition result of the at least one first other surface image according to the image recognition model, and recognize the second other surface image associated with other target surface images in a certain target surface image subsequence according to the second recognition result to obtain a third recognition result;

[0021] The determination module is configured to determine whether the line segment to be detected is abnormal based on the second recognition result and the third recognition result, corresponding to each target surface image in the subsequence of a certain target surface image.

[0022] Thirdly, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of a defect detection method for lines in a power distribution network according to any embodiment of the present invention.

[0023] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor performs the steps of a defect detection method for lines in a power distribution network according to any embodiment of the present invention.

[0024] The present application discloses a method and system for defect detection of lines in a power distribution network. This method compares various surface images within the same surface image set, selects a target surface image from the set based on the comparison results, constructs a target surface image sequence, and divides the sequence according to a preset truncation rule to obtain at least one target surface image subsequence. Any target surface image is then extracted from a subsequence, and the image is recognized using a preset image recognition model. The system then determines whether the first recognition result output by the image recognition model is abnormal. This effectively improves the recognition efficiency of the line under inspection when dealing with continuous scratches. Attached Figure Description

[0025] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 A flowchart illustrating a defect detection method for lines in a power distribution network, provided as an embodiment of the present invention;

[0027] Figure 2 The present invention provides a structural block diagram of a defect detection system for lines in a power distribution network according to an embodiment of the present invention;

[0028] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] Please see Figure 1 The diagram shows a flowchart of a defect detection method for lines in a power distribution network according to this application.

[0031] like Figure 1 As shown, the defect detection method for lines in a power distribution network specifically includes the following steps:

[0032] Step S101: Obtain at least one set of surface images of a line segment to be detected, wherein the set of surface images contains surface images in at least one direction.

[0033] In this step, the direction refers to the direction in which the line segment to be inspected is photographed. For example, for a line segment to be inspected, the shooting directions are directly above, directly in front, directly below, and directly behind, so that the entire outer surface of the line segment to be inspected can be photographed.

[0034] Step S102: Compare the surface images in the same surface image set, select the target surface image in the same surface image set according to the comparison result, and construct the association relationship between the target surface image and other surface images, wherein the other surface images are the surface images in the same surface image set excluding the target surface image.

[0035] In this step, the similarity between various surface images in the same surface image set is obtained; the surface image with the lowest similarity to other surface images is selected as the target surface image.

[0036] It should be noted that similarity can be calculated by using the VGGNet network model to perform image recognition on the two processed surface images, obtaining the first feature and the second feature corresponding to the two surface images respectively. Then, the difference between the first feature and the second feature is calculated, and the difference is subtracted from 1 to obtain the similarity between the two surface images.

[0037] The VGGNet network model is obtained through machine learning using multiple sets of training and detection images.

[0038] Step S103: Sort the target surface images according to the location information of the at least one line segment to be detected on the preset electronic map to obtain a target surface image sequence.

[0039] Step S104: Divide the target surface image sequence according to the preset cropping rules to obtain at least one target surface image subsequence, wherein each target surface image in a target surface image subsequence belongs to the same direction.

[0040] In this step, the orientation of each target surface image in the target surface image sequence is obtained; at least one consecutive target surface image with the same orientation is divided into the same target surface image subsequence to obtain at least one target surface image subsequence.

[0041] Step S105: Extract any target surface image from a certain target surface image subsequence, identify the target surface image based on a preset image recognition model, and determine whether the first recognition result output by the image recognition model is abnormal.

[0042] In one specific implementation, if the first identification result is abnormal, the corresponding line segments to be detected in each target surface image subsequence are directly defined as abnormal line segments. Since line damage is continuous, that is, if the current line segment to be detected has scratches, the next or previous line segment to be detected adjacent to the current line segment to be detected may also have scratches. Therefore, the method of this implementation can identify the line segments to be detected relatively quickly.

[0043] Step S106: If the first recognition result is not abnormal, then at least one first other surface image associated with any of the target surface images is obtained, a second recognition result of the at least one first other surface image is obtained according to the image recognition model, and a second other surface image associated with other target surface images in a certain target surface image subsequence is recognized according to the second recognition result to obtain a third recognition result.

[0044] In this step, a first other surface image and a second other surface image in the same direction are associated to obtain a target association relationship; a second recognition result of a certain first other surface image is obtained; and a third recognition result of the certain first other surface image is obtained by assigning the second recognition result to a certain first other surface image that has a target association relationship with the first other surface image.

[0045] In one specific implementation, to improve the accuracy of identification, all other surface images can be identified sequentially. Compared to the implementation method in step S106, although the identification efficiency is slightly reduced, the identification accuracy is better, making it a feasible implementation method. For example, for questionable sections of the line to be detected, the method of sequentially identifying all other surface images can be used.

[0046] Step S107: Determine whether the line segment to be detected is abnormal based on the second recognition result and the third recognition result, corresponding to each target surface image in the target surface image subsequence.

[0047] In this step, it is determined whether the second identification result is abnormal; if the second identification result is abnormal, the line segment to be detected corresponding to the at least one first other surface image is determined to be an abnormal line segment, otherwise it is a normal line segment; it is determined whether the third identification result is abnormal; if the third identification result is abnormal, the line segment to be detected corresponding to the at least one second other surface image is determined to be an abnormal line segment, otherwise it is a normal line segment.

[0048] In summary, the method of this application compares various surface images in the same surface image set, selects a target surface image from the same surface image set based on the comparison results, constructs a target surface image sequence, divides the target surface image sequence based on a preset truncation rule to obtain at least one target surface image subsequence, extracts any target surface image from a target surface image subsequence, identifies any target surface image based on a preset image recognition model, and determines whether the first recognition result output by the image recognition model is abnormal. In this way, when facing the situation where there are continuous scratches on the line to be detected, the recognition efficiency of the line to be detected can be effectively improved.

[0049] Please see Figure 2 The diagram shows a structural block diagram of a defect detection system for lines in a power distribution network according to this application.

[0050] like Figure 2 As shown, the fault monitoring system 200 includes an acquisition module 210, a construction module 220, a sorting module 230, a division module 240, a first identification module 250, a second identification module 260, and a determination module 270.

[0051] The acquisition module 210 is configured to acquire at least one set of surface images of a line segment to be detected, wherein the set of surface images includes surface images in at least one direction; the construction module 220 is configured to compare the surface images in the same set of surface images, select a target surface image in the same set of surface images based on the comparison results, and construct the association relationship between the target surface image and other surface images, wherein the other surface images are the surface images in the same set of surface images excluding the target surface image; the sorting module 230 is configured to sort the target surface images according to the location information of the at least one line segment to be detected on a preset electronic map to obtain a target surface image sequence; the segmentation module 240 is configured to segment the target surface image sequence based on a preset truncation rule to obtain at least one target surface image subsequence, wherein each target surface image in a target surface image subsequence... All belong to the same direction; the first identification module 250 is configured to extract any target surface image from a certain target surface image subsequence, identify the any target surface image based on a preset image recognition model, and determine whether the first identification result output by the image recognition model is abnormal; the second identification module 260 is configured to, if the first identification result is not abnormal, obtain at least one first other surface image associated with the any target surface image, obtain the second identification result of the at least one first other surface image according to the image recognition model, and identify the second other surface image associated with other target surface images in the certain target surface image subsequence according to the second identification result to obtain a third identification result; the determination module 270 is configured to determine whether the line segment to be detected corresponding to each target surface image in the certain target surface image subsequence is abnormal according to the second identification result and the third identification result.

[0052] It should be understood that Figure 2 The modules and references described in the document Figure 1 The steps described in the text correspond to those in the method described above. Therefore, the operations, features, and corresponding technical effects described above also apply to the method described in the text. Figure 2 The various modules in the document will not be described in detail here.

[0053] In other embodiments, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor performs the defect detection method for lines in a power distribution network as described in any of the above method embodiments.

[0054] In one embodiment, the computer-readable storage medium of the present invention stores computer-executable instructions, which are configured as follows:

[0055] Acquire a surface image set of at least one line segment to be detected, wherein the surface image set contains surface images in at least one direction;

[0056] Compare the surface images in the same surface image set, select the target surface image in the same surface image set according to the comparison results, and construct the association relationship between the target surface image and other surface images, wherein the other surface images are the surface images in the same surface image set excluding the target surface image;

[0057] Based on the location information of the at least one line segment to be detected on a preset electronic map, sort the target surface images to obtain a target surface image sequence.

[0058] The target surface image sequence is divided according to a preset cropping rule to obtain at least one target surface image subsequence, wherein each target surface image in a target surface image subsequence belongs to the same direction;

[0059] Extract any target surface image from a subsequence of target surface images, identify the target surface image based on a preset image recognition model, and determine whether the first recognition result output by the image recognition model is abnormal.

[0060] If the first recognition result is not abnormal, then at least one first other surface image associated with any of the target surface images is obtained, a second recognition result of the at least one first other surface image is obtained according to the image recognition model, and a second other surface image associated with other target surface images in a certain target surface image subsequence is recognized according to the second recognition result to obtain a third recognition result;

[0061] Based on the second and third identification results, it is determined whether the line segment to be detected is abnormal, corresponding to each target surface image in the subsequence of a certain target surface image.

[0062] Computer-readable storage media may include a stored program area and a stored data area, wherein the stored program area may store an operating system and an application program required for at least one function; the stored data area may store data created based on the use of a defect detection system for lines in a power distribution network. Furthermore, the computer-readable storage medium may include high-speed random access memory, and may also include memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include memory remotely disposed relative to a processor, which can be connected via a network to a defect detection system for lines in a power distribution network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0063] Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present invention, such as... Figure 3 As shown, the device includes a processor 310 and a memory 320. The electronic device may also include an input device 330 and an output device 340. The processor 310, memory 320, input device 330, and output device 340 can be connected via a bus or other means. Figure 3 Taking a bus connection as an example, the memory 320 is the computer-readable storage medium described above. The processor 310 executes various server functions and data processing by running non-volatile software programs, instructions, and modules stored in the memory 320, thereby implementing the defect detection method for lines in a power distribution network as described in the above method embodiment. The input device 330 can receive input digital or character information and generate key signal inputs related to user settings and function control of the defect detection system for lines in a power distribution network. The output device 340 may include a display screen or other display device.

[0064] The aforementioned electronic device can execute the method provided in the embodiments of the present invention, and has the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in the embodiments of the present invention.

[0065] In one implementation, the above-described electronic device is applied to a defect detection system for lines in a power distribution network, for a client, and includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to:

[0066] Acquire a surface image set of at least one line segment to be detected, wherein the surface image set contains surface images in at least one direction;

[0067] Compare the surface images in the same surface image set, select the target surface image in the same surface image set according to the comparison results, and construct the association relationship between the target surface image and other surface images, wherein the other surface images are the surface images in the same surface image set excluding the target surface image;

[0068] Based on the location information of the at least one line segment to be detected on a preset electronic map, sort the target surface images to obtain a target surface image sequence.

[0069] The target surface image sequence is divided according to a preset cropping rule to obtain at least one target surface image subsequence, wherein each target surface image in a target surface image subsequence belongs to the same direction;

[0070] Extract any target surface image from a subsequence of target surface images, identify the target surface image based on a preset image recognition model, and determine whether the first recognition result output by the image recognition model is abnormal.

[0071] If the first recognition result is not abnormal, then at least one first other surface image associated with any of the target surface images is obtained, a second recognition result of the at least one first other surface image is obtained according to the image recognition model, and a second other surface image associated with other target surface images in a certain target surface image subsequence is recognized according to the second recognition result to obtain a third recognition result;

[0072] Based on the second and third identification results, it is determined whether the line segment to be detected is abnormal, corresponding to each target surface image in the subsequence of a certain target surface image.

[0073] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0074] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for defect detection in power distribution lines, characterized in that, include: Acquire a surface image set of at least one line segment to be detected, wherein the surface image set contains surface images in at least one direction; Compare the surface images in the same surface image set, select the target surface image in the same surface image set according to the comparison results, and construct the association relationship between the target surface image and other surface images, wherein the other surface images are the surface images in the same surface image set excluding the target surface image; Based on the location information of the at least one line segment to be detected on a preset electronic map, sort the target surface images to obtain a target surface image sequence. The target surface image sequence is divided according to a preset cropping rule to obtain at least one target surface image subsequence, wherein each target surface image in a target surface image subsequence belongs to the same direction; Extract any target surface image from a subsequence of target surface images, identify the target surface image based on a preset image recognition model, and determine whether the first recognition result output by the image recognition model is abnormal. If the first recognition result is not abnormal, then at least one first other surface image associated with any of the target surface images is obtained, a second recognition result of the at least one first other surface image is obtained according to the image recognition model, and a second other surface image associated with other target surface images in a certain target surface image subsequence is recognized according to the second recognition result to obtain a third recognition result; Based on the second and third identification results, it is determined whether the line segment to be detected is abnormal, corresponding to each target surface image in the subsequence of a certain target surface image.

2. The defect detection method for lines in a power distribution network according to claim 1, characterized in that, The step of comparing various surface images in the same surface image set and selecting a target surface image in the same surface image set based on the comparison results includes: Obtain the similarity between surface images in the same surface image set; The surface image with the lowest similarity to other surface images is selected as the target surface image.

3. The defect detection method for lines in a power distribution network according to claim 1, characterized in that, The process of dividing the target surface image sequence based on preset truncation rules to obtain at least one target surface image sub-sequence includes: Obtain the orientation of each target surface image in the target surface image sequence; At least one target surface image that is continuous and has the same orientation is divided into the same target surface image subsequence to obtain at least one target surface image subsequence.

4. The defect detection method for lines in a power distribution network according to claim 1, characterized in that, After determining whether the first recognition result output by the image recognition model is abnormal, the method further includes: If the first identification result is abnormal, then the line segment to be detected corresponding to each target surface image in the subsequence of a certain target surface image is directly defined as an abnormal line segment.

5. A defect detection method for lines in a power distribution network according to claim 1, characterized in that, The step of identifying a second other surface image associated with other target surface images in the target surface image subsequence based on the second identification result to obtain a third identification result includes: By associating the first other surface image and the second other surface image in the same direction, the target association relationship is obtained; A second recognition result of a first other surface image is obtained, and a third recognition result of the first other surface image is obtained by assigning the second recognition result to a first other surface image that has a target association relationship with the first other surface image.

6. A defect detection method for lines in a power distribution network according to claim 1, characterized in that, The step of determining whether the line segment to be detected is abnormal based on the second recognition result and the third recognition result and corresponding to each target surface image in the subsequence of a certain target surface image includes: Determine whether the second identification result is abnormal; If the second identification result is abnormal, the line segment to be detected corresponding to the at least one first other surface image is determined to be an abnormal line segment; otherwise, it is a normal line segment. Determine whether the third identification result is abnormal; If the third identification result is abnormal, the line segment to be detected corresponding to at least one second other surface image is determined to be an abnormal line segment; otherwise, it is a normal line segment.

7. A defect detection system for lines in a power distribution network, characterized in that, include: The acquisition module is configured to acquire at least one set of surface images of a line segment to be detected, wherein the set of surface images includes surface images in at least one direction; The construction module is configured to compare various surface images in the same surface image set, select a target surface image in the same surface image set according to the comparison results, and construct the association relationship between the target surface image and other surface images, wherein the other surface images are surface images in the same surface image set excluding the target surface image; The sorting module is configured to sort each target surface image according to the location information of the at least one line segment to be detected on a preset electronic map, so as to obtain a target surface image sequence. The segmentation module is configured to segment the target surface image sequence based on a preset truncation rule to obtain at least one target surface image subsequence, wherein each target surface image in a target surface image subsequence belongs to the same direction; The first recognition module is configured to extract any target surface image from a subsequence of target surface images, recognize the any target surface image based on a preset image recognition model, and determine whether the first recognition result output by the image recognition model is abnormal. The second recognition module is configured to, if the first recognition result is not abnormal, acquire at least one first other surface image associated with any target surface image, acquire a second recognition result of the at least one first other surface image according to the image recognition model, and recognize the second other surface image associated with other target surface images in a certain target surface image subsequence according to the second recognition result to obtain a third recognition result; The determination module is configured to determine whether the line segment to be detected is abnormal based on the second recognition result and the third recognition result, corresponding to each target surface image in the subsequence of a certain target surface image.

8. An electronic device, characterized in that, include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, it implements the method described in any one of claims 1 to 6.

Images