A method and device for detecting live voltage of crimping hardware

Abstract

The present application provides a method and device for detecting live crimping hardware, including: obtaining a transmission line inspection image, and when a crimping hardware is detected from the transmission line inspection image, extracting the appearance contour features of the crimping hardware by image feature extraction; dividing the crimping hardware into a number of structural segments according to the appearance contour features; calculating the structural proportions of each structural segment in the crimping hardware according to the pixel size occupied by each structural segment, and obtaining the process detection results of the crimping hardware based on the comparison result of the preset ratio reference range and the structural proportion. The present application uses a drone to take images of crimping hardware assembled on a transmission line, and then uses an image feature recognition method to determine the ratio between the crimping area and the non-crushing area of ​​the crimping hardware, and identifies the crimping process, which solves the technical problems of the existing crimping hardware detection method being time-consuming and having low safety guarantees for inspection personnel.

Description

Technical Field

[0001] The present application relates to the technical field of electric power inspection, and in particular to a method for detecting live crimping hardware. Background Art

[0002] At present, crimping fittings are the main means to achieve uninterrupted long-distance power transmission of UHV transmission lines. Commonly used crimping fittings include tension clamps and joint pipes. With the rapid development of power grids, voltage levels are getting higher and higher, and transmission lines are becoming more and more dense. The use of crimping fittings on transmission lines has also increased significantly, making the performance defects of crimping fittings an important factor affecting the safe and stable operation of power grids.

[0003] Since crimping is a hidden project, there are few methods to identify the quality of crimping during acceptance and operation, and the operability is not high. The existing detection method mainly uses X-ray flaw detectors for non-destructive testing to determine slag cracks, pores and poor welding. However, due to the complex environment of power transmission lines and inconvenient transportation conditions, it is often necessary to rely on manpower to carry equipment for line inspection, which is time-consuming and has low safety guarantees for inspection personnel. Summary of the invention

[0004] The present application provides a method and device for detecting live crimping hardware, which are used to solve the technical problems that the existing crimping hardware detection method is time-consuming and has low safety guarantee for inspectors.

[0005] The first aspect of the present application provides a method for detecting live crimping hardware, comprising:

[0006] Acquire a power transmission line inspection image, and when a crimping fitting is detected from the power transmission line inspection image, extract the appearance contour features of the crimping fitting by using an image feature extraction method;

[0007] According to the appearance contour features, the crimping hardware is divided into a plurality of structural segments;

[0008] According to the pixel size occupied by each of the structural segments, the structural ratio of each structural segment in the crimping hardware is calculated, and based on the comparison result between a preset ratio reference range and the structural ratio, the process detection result of the crimping hardware is obtained.

[0009] Optionally, before extracting the appearance contour features of the crimping hardware by image feature extraction, the method further includes:

[0010] Based on the position of the crimping fitting in the power transmission line inspection image, the power transmission line inspection image is cropped to obtain the crimping fitting image.

[0011] Optionally, after extracting the appearance contour features of the crimping hardware by image feature extraction, the method further includes:

[0012] According to the appearance contour feature, the slope of the crimping hardware is determined by a Hough transform detection method, wherein the slope of the crimping hardware is specifically the slope of a line connecting two ends of the crimping hardware in the image;

[0013] When the slope of the crimping hardware exceeds the preset slope threshold range, the image is rotated according to the calculated rotation angle until the slope of the crimping hardware is within the slope threshold range, wherein the rotation angle is calculated based on the difference between the slope of the crimping hardware and the slope threshold range.

[0014] Optionally, the acquiring of the transmission line inspection image, when a crimping fitting is detected from the transmission line inspection image, extracting the appearance contour features of the crimping fitting by image feature extraction specifically includes:

[0015] Acquire transmission line inspection images;

[0016] The transmission line inspection image is input into a preset target detection model. When the target detection model detects crimping hardware from the transmission line inspection image, the appearance contour features of the crimping hardware are extracted by image feature extraction.

[0017] Optionally, the target detection model is constructed as follows:

[0018] The preset crimping hardware sample images are input into the neural network model for training to obtain the trained target detection model, wherein the crimping hardware sample images include: tension clamp sample images, crimping tube sample images and connecting tube sample images.

[0019] The second aspect of the present application provides a device for detecting live crimping hardware, comprising:

[0020] A feature extraction unit is used to obtain a transmission line inspection image, and when a crimping hardware is detected from the transmission line inspection image, an appearance contour feature of the crimping hardware is extracted by an image feature extraction method;

[0021] A structural segmentation unit, used to divide the crimping hardware into a plurality of structural segments according to the appearance contour features;

[0022] The detection unit is used to calculate the structural proportion of each structural segment in the crimping hardware according to the pixel size occupied by each structural segment, and obtain the process detection result of the crimping hardware based on the comparison result between the preset proportion reference range and the structural proportion.

[0023] Optionally, it also includes:

[0024] A cropping unit is used to crop the power transmission selected inspection image based on the position of the crimping fitting in the power transmission line inspection image to obtain the crimping fitting image.

[0025] Optionally, it also includes:

[0026] A slope calculation unit, used to determine the slope of the crimping hardware by Hough transform detection according to the appearance contour feature, wherein the slope of the crimping hardware is specifically the slope of the line connecting the two ends of the crimping hardware in the image;

[0027] An image rotation unit is used to rotate the image according to a calculated rotation angle when the slope of the crimping hardware exceeds a preset slope threshold range, until the slope of the crimping hardware is within the slope threshold range, wherein the rotation angle is calculated based on the difference between the slope of the crimping hardware and the slope threshold range.

[0028] Optionally, the feature extraction unit specifically includes:

[0029] An image acquisition subunit, used for acquiring transmission line inspection images;

[0030] The feature extraction subunit is used to input the transmission line inspection image into a preset target detection model. When the target detection model detects crimping hardware from the transmission line inspection image, the appearance contour features of the crimping hardware are extracted by image feature extraction.

[0031] Optionally, it also includes:

[0032] The model training unit is used to input preset crimping hardware sample images into the neural network model for training to obtain the trained target detection model, wherein the crimping hardware sample images include: tension clamp sample images, crimping tube sample images and connecting tube sample images.

[0033] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0034] The present application provides a method for detecting live crimping hardware, comprising: acquiring a transmission line inspection image, and when a crimping hardware is detected from the transmission line inspection image, extracting the appearance contour features of the crimping hardware by means of image feature extraction; dividing the crimping hardware into a plurality of structural segments according to the appearance contour features; calculating the structural proportion of each structural segment in the crimping hardware according to the pixel size occupied by each structural segment, and obtaining a process inspection result of the crimping hardware based on a comparison result between a preset ratio reference range and the structural proportion.

[0035] This application combines drone inspection technology, uses drones to capture images of crimping hardware installed on transmission lines, and then uses image feature recognition methods to determine the proportional relationship between the crimping area and the non-crushing area of ​​the crimping hardware to identify the crimping process. This can effectively reduce the cost of construction acceptance and hidden danger inspection of crimping hardware, as well as the risk of personnel carrying equipment for field operations, and solves the technical problems of existing crimping hardware inspection methods that are time-consuming and have low safety guarantees for inspection personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0037] Figure 1 A schematic diagram of a flow chart of a first embodiment of a method for detecting live crimping hardware provided in the present application;

[0038] Figure 2 A schematic flow chart of a second embodiment of a method for detecting live crimping hardware provided by the present application;

[0039] Figure 3 This is a schematic structural diagram of a first embodiment of a device for detecting live crimping hardware provided in the present application.

[0040] Figure 4 This is a schematic diagram of the structural segment division of the crimping hardware provided in this application. DETAILED DESCRIPTION

[0041] The embodiments of the present application provide a method and device for detecting live crimping hardware, which are used to solve the technical problems that the existing crimping hardware detection method is time-consuming and has low safety guarantee for inspection personnel.

[0042] In order to make the purpose, features, and advantages of the invention of this application more obvious and easy to understand, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the embodiments described below are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0043] See also Figure 1 The first embodiment of the present application provides a method for detecting live crimping hardware, comprising:

[0044] Step 101: acquiring a power transmission line inspection image. When a crimping fitting is detected from the power transmission line inspection image, an appearance contour feature of the crimping fitting is extracted by an image feature extraction method.

[0045] It should be noted that first, the transmission line inspection image is obtained, specifically, the refined transmission line inspection image data taken by a drone. Then, when the crimping hardware to be inspected is detected from the captured image, the appearance contour image features of the crimping hardware in the inspection image are extracted by image feature extraction.

[0046] Step 102, dividing the crimping fitting into a plurality of structural segments according to the appearance contour features;

[0047] Next, according to the appearance contour features of the crimping hardware extracted in the above steps, the crimping hardware in the image is divided into several structural segments. The specific division can be referred to Figure 4 shown.

[0048] Step 103: Calculate the structural proportion of each structural segment in the crimping hardware according to the pixel size occupied by each structural segment, and obtain the process detection result of the crimping hardware based on the comparison result between the preset proportion reference range and the structural proportion.

[0049] Next, according to the pixel size occupied by each structural segment, the ratio of the compression area to the non-compression area of ​​the crimping hardware is calculated, and based on the comparison result of the preset ratio reference value and the ratio, the process inspection result of the crimping hardware is obtained to determine whether the process of the crimping tube meets the standards.

[0050] This embodiment combines drone inspection technology, uses drones to capture images of crimping hardware installed on transmission lines, and then uses image feature recognition methods to determine the proportional relationship between the crimping area and the non-crushing area of ​​the crimping hardware to identify the crimping process. This can effectively reduce the cost of construction acceptance and hidden danger inspection of crimping hardware, as well as the risk of personnel carrying equipment for field operations, and solves the technical problems of existing crimping hardware inspection methods that are time-consuming and have low safety guarantees for inspection personnel.

[0051] The above is a detailed description of the first embodiment of a method for detecting live crimping hardware provided by the present application. The following is a detailed description of the second embodiment of a method for detecting live crimping hardware provided by the present application.

[0052] See also Figure 2 The second embodiment of the present application provides a method for detecting live crimping hardware, comprising:

[0053] Step 201, obtaining a transmission line inspection image;

[0054] Step 202: input the power transmission line inspection image into a preset target detection model. When the target detection model detects crimping fittings from the power transmission line inspection image, subsequent steps are executed.

[0055] Among them, in order to ensure the recognition accuracy of crimping hardware, this embodiment adopts a deep learning target detection algorithm, and inputs the preset crimping hardware sample images into the neural network model for training to obtain a trained target detection model, wherein the crimping hardware sample images include: tension clamp sample images, crimping tube sample images and connecting tube sample images. The specific target detection model recognition mechanism is as follows:

[0056] A. The image is passed through the VGG16 feature extraction network to obtain a series of feature blocks (feature maps), and these feature maps share parameters with the subsequent RPN (region proposal network) and FastR-CNN network.

[0057] B. The features obtained in the previous step are input into the RPN network to obtain a series of region proposal boxes. The RPN network is trained with the training set to enable the network to predict the target proposal box, that is, to use a neural network instead of the selective search algorithm (selective search) of traditional image processing to predict and generate the proposed region.

[0058] C. After obtaining a series of region proposal boxes, the semantic features corresponding to these region proposal boxes are mapped through the region of interest pooling layer (ROI pooling) to output feature information of the same size, and then these feature information are input into the final fully connected layer.

[0059] D. The output of the fully connected layer obtains accurate category prediction and target positioning through the classification function softmax and the bounding box (bbox) regression function. Finally, the predicted targets of similar positions and the same category are eliminated through the non-maximum suppression (NMS NMS) algorithm to obtain the final target prediction box.

[0060] Step 203: based on the position of the crimping fittings in the power transmission line inspection image, the power transmission line inspection image is cropped to obtain a crimping fitting image.

[0061] Among them, after the crimping hardware is detected and its position in the image is identified, in order to improve the recognition accuracy of the crimping hardware in the next step, the crimping hardware can be cropped out of the original image to obtain a fine image containing only the crimping hardware.

[0062] Step 204: extracting the appearance contour features of the crimping hardware by image feature extraction.

[0063] It should be noted that first, the transmission line inspection image is obtained, specifically, the refined transmission line inspection image data taken by a drone. Then, when the crimping hardware to be inspected is detected from the captured image, the appearance contour image features of the crimping hardware in the inspection image are extracted by image feature extraction.

[0064] Step 205: Divide the crimping fitting into a plurality of structural segments according to the appearance contour features;

[0065] Next, according to the appearance contour features of the crimping hardware extracted in the above step, the crimping hardware in the image is divided into a plurality of structural segments.

[0066] Step 206: Determine the slope of the crimping fitting by Hough transform detection according to the appearance contour feature. The slope of the crimping fitting is specifically the slope of the line connecting the two ends of the crimping fitting in the image.

[0067] Step 207: When the slope of the crimping fitting exceeds the preset slope threshold range, the image is rotated according to the calculated rotation angle until the slope of the crimping fitting is within the slope threshold range, wherein the rotation angle is calculated based on the difference between the slope of the crimping fitting and the slope threshold range.

[0068] Among them, in the actual application process, the position of the crimping hardware may be tilted or deflected due to the angle of the drone shooting. In order to make the identification of various parts of the crimping hardware more accurate, the rotation angle detection based on Hough transform is adopted in this process. The slope of the crimping hardware and the boundary is calculated through the Hough transform straight line detection algorithm. The rotation angle is calculated through the trigonometric function inversion, and the image is rotated so that the two ends of the crimping hardware are in a horizontal position, thereby further improving the accuracy of the detection.

[0069] Step 208: Calculate the structural proportion of each structural segment in the crimping hardware according to the pixel size occupied by each structural segment, and obtain the process detection result of the crimping hardware based on the comparison result between the preset proportion reference range and the structural proportion.

[0070] Next, according to the pixel size occupied by each structural segment, the ratio of the compression area to the non-compression area of ​​the crimping hardware is calculated, and based on the comparison result of the preset ratio reference value and the ratio, the process inspection result of the crimping hardware is obtained to determine whether the process of the crimping tube meets the standards.

[0071] The above is a detailed description of the second embodiment of a method for detecting live crimping hardware provided by the present application. The following is a detailed description of the first embodiment of a device for detecting live crimping hardware provided by the present application.

[0072] See also Figure 3 The third embodiment of the present application provides a device for detecting a live crimping hardware, comprising:

[0073] The feature extraction unit 301 is used to obtain a transmission line inspection image. When a crimping fitting is detected from the transmission line inspection image, the appearance contour features of the crimping fitting are extracted by image feature extraction.

[0074] The structural segmentation unit 302 is used to divide the crimping fitting into a plurality of structural segments according to the appearance contour characteristics;

[0075] The detection unit 303 is used to calculate the structural proportion of each structural segment in the crimping hardware according to the pixel size occupied by each structural segment, and obtain the process detection result of the crimping hardware based on the comparison result between the preset proportion reference range and the structural proportion.

[0076] Furthermore, it also includes:

[0077] The cropping unit 304 is used to crop the power transmission selected inspection image based on the position of the crimping fittings in the power transmission line inspection image to obtain the crimping fittings image.

[0078] Furthermore, it also includes:

[0079] The slope calculation unit 305 is used to determine the slope of the crimping hardware according to the appearance contour features, and the straight line slope is specifically the slope of the line connecting the two ends of the crimping hardware in the image;

[0080] The image rotation unit 306 is used to rotate the image by Hough transform detection when the slope of the crimping fitting exceeds a preset slope threshold range, until the slope of the crimping fitting is within the slope threshold range.

[0081] Furthermore, the feature extraction unit 301 specifically includes:

[0082] The image acquisition subunit 3011 is used to acquire the transmission line inspection image;

[0083] The feature extraction subunit 3012 is used to input the transmission line inspection image into a preset target detection model. When the target detection model detects crimping hardware from the transmission line inspection image, the appearance contour features of the crimping hardware are extracted by image feature extraction.

[0084] Furthermore, it also includes:

[0085] The model training unit 307 is used to input preset crimping hardware sample images into the neural network model for training to obtain a trained target detection model, wherein the crimping hardware sample images include: tension clamp sample images, crimping tube sample images and connecting tube sample images.

[0086] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0087] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0088] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein, for example. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0089] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0090] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0091] If the integrated unit is implemented in the form of a software functional 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 solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0092] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for detecting live crimping hardware, characterized in that: include: Inputting preset crimping fitting sample images into a neural network model for training to obtain a trained target detection model, wherein the crimping fitting sample images include: a tension clamp sample image, a crimping tube sample image, and a connecting tube sample image; Acquire a power transmission line inspection image, and when a crimping fitting is detected from the power transmission line inspection image, extract the appearance contour features of the crimping fitting by using an image feature extraction method; According to the appearance contour feature, the slope of the crimping hardware is determined by a Hough transform detection method, wherein the slope of the crimping hardware is specifically the slope of a line connecting two ends of the crimping hardware in the image; When the slope of the crimping fitting exceeds a preset slope threshold range, the image is rotated according to the calculated rotation angle until the slope of the crimping fitting is within the slope threshold range, wherein the rotation angle is calculated based on the difference between the slope of the crimping fitting and the slope threshold range; According to the appearance contour features, the crimping hardware is divided into a plurality of structural segments; According to the pixel size occupied by each of the structural segments, the structural ratio of each structural segment in the crimping hardware is calculated, and based on the comparison result between a preset ratio reference range and the structural ratio, the process detection result of the crimping hardware is obtained.

2. A method for detecting live crimping hardware according to claim 1, characterized in that: Before extracting the appearance contour features of the crimping hardware by image feature extraction, the method further includes: Based on the position of the crimping fitting in the power transmission line inspection image, the power transmission line inspection image is cropped to obtain the crimping fitting image.

3. A method for detecting live crimping hardware according to claim 1, characterized in that: The acquiring of the transmission line inspection image, when the crimping fittings are detected from the transmission line inspection image, extracting the appearance contour features of the crimping fittings by image feature extraction specifically includes: Acquire transmission line inspection images; The transmission line inspection image is input into a preset target detection model. When the target detection model detects crimping hardware from the transmission line inspection image, the appearance contour features of the crimping hardware are extracted by image feature extraction.

4. A device for detecting live crimping hardware, characterized in that: include: A model training unit is used to input preset crimping fitting sample images into a neural network model for training to obtain a trained target detection model, wherein the crimping fitting sample images include: a tension clamp sample image, a crimping tube sample image, and a connecting tube sample image; A feature extraction unit is used to obtain a transmission line inspection image, and when a crimping hardware is detected from the transmission line inspection image, an appearance contour feature of the crimping hardware is extracted by an image feature extraction method; A slope calculation unit, used to determine the slope of the crimping hardware by Hough transform detection according to the appearance contour feature, wherein the slope of the crimping hardware is specifically the slope of the line connecting the two ends of the crimping hardware in the image; An image rotation unit, used to rotate the image according to a calculated rotation angle when the slope of the crimping fitting exceeds a preset slope threshold range, until the slope of the crimping fitting is within the slope threshold range, wherein the rotation angle is calculated based on the difference between the slope of the crimping fitting and the slope threshold range; A structural segmentation unit, used to divide the crimping hardware into a plurality of structural segments according to the appearance contour features; The detection unit is used to calculate the structural proportion of each structural segment in the crimping hardware according to the pixel size occupied by each structural segment, and obtain the process detection result of the crimping hardware based on the comparison result between the preset proportion reference range and the structural proportion.

5. A device for detecting live pressure fittings according to claim 4, characterized in that: Also includes: A cropping unit is used to crop the transmission line inspection image based on the position of the crimping fitting in the transmission line inspection image to obtain the crimping fitting image.

6. A device for detecting live pressure fittings according to claim 4, characterized in that: The feature extraction unit specifically includes: An image acquisition subunit, used for acquiring transmission line inspection images; The feature extraction subunit is used to input the transmission line inspection image into a preset target detection model. When the target detection model detects crimping hardware from the transmission line inspection image, the appearance contour features of the crimping hardware are extracted by image feature extraction.

Images