A method and related device for marking transmission line images

Abstract

The present application discloses a method for annotating a power transmission line image and a related device, the method comprising: obtaining multi-angle images of a target line; performing shooting classification on the multi-angle images according to the EXIF ​​information extracted from the multi-angle images to obtain a classified line image, wherein the EXIF ​​information includes shooting posture and sensitivity; performing texture feature extraction operation on the classified line image using a preset LBP algorithm to obtain a texture feature image; constructing a labeled annotation layer according to the multi-angle images and the texture feature image; generating a preset polygonal block based on the annotation layer, wherein the preset polygonal block includes an angle label and a texture label; and annotating the image corresponding to the target line according to the preset polygonal block. The present application can solve the technical problem that the annotation operation of existing image data is time-consuming and inefficient, resulting in poor data management results.

Description

Technical Field

[0001] The present application relates to the technical field of image annotation, and in particular to a method for annotating images of power transmission lines and related devices. Background Art

[0002] With the gradual and in-depth application of computer vision and artificial intelligence technology in the field of power grid inspection, the labeling of artificial intelligence data sets in the field of power grid inspection has become a new standard work. Inspectors have reduced their work outside and instead spend part of their time teaching machines how to identify equipment on power grid transmission lines and identify related defects. With the gradual maturity of acquisition equipment and the gradual application of related algorithms, a large number of data sets are facing heavy labeling and maintenance management work. In particular, some special data sets have new requirements and challenges for the labeling of raw data.

[0003] Most intelligent recognition uses CNN, RCNN, and Fast-RCNN to conduct deep learning training on power transmission line equipment, which relies on a large number of data sets, and the data sets need to be annotated before use; especially the use of GAN (Generative Adversarial Network) architecture to generate and identify defective images of power transmission line equipment continues to expand, and the lagging traditional image annotation tools cannot provide the corresponding data format. In general, the image annotation and data management efficiency for power transmission line inspection is not high at present, mainly relying on operators to hand-draw using annotation software. Often, the same target object needs to process multiple images, which is time-consuming, inefficient, and not friendly to the annotators. Summary of the invention

[0004] The present application provides a power transmission line image annotation method and related devices, which are used to solve the technical problem that the existing image data annotation operation is time-consuming and inefficient, resulting in poor data management results.

[0005] In view of this, the first aspect of the present application provides a method for annotating a power transmission line image, comprising:

[0006] Acquire multi-angle images of the target line;

[0007] Classify the multi-shooting angle images according to the EXIF ​​information extracted from the multi-shooting angle images to obtain classified line images, wherein the EXIF ​​information includes shooting posture and sensitivity;

[0008] Using a preset LBP algorithm to perform a texture feature extraction operation on the classified line image to obtain a texture feature image;

[0009] Constructing a labeled annotation layer according to the multi-shooting angle images and the texture feature images;

[0010] Generate a preset polygonal block based on the annotation layer, wherein the preset polygonal block includes an angle label and a texture label;

[0011] The image corresponding to the target line is labeled according to the preset polygonal block.

[0012] Preferably, constructing a labeled annotation layer according to the multi-shooting angle images and the texture feature image includes:

[0013] Constructing an initial annotation layer according to the multi-shooting angle images and the texture feature images;

[0014] Angle labels and texture labels are added to the angle layer and texture feature layer in the initial annotation layer respectively to obtain an annotation layer.

[0015] Preferably, the step of constructing a labeled annotation layer according to the multi-shooting angle images and the texture feature images further includes:

[0016] Performing grayscale processing on the multi-shooting angle images to obtain grayscale images;

[0017] Accordingly, constructing a labeled annotation layer according to the multi-shooting angle image and the texture feature image includes:

[0018] A labeled annotation layer is constructed according to the multi-shooting angle images, the texture feature images and the grayscale images.

[0019] Preferably, the step of labeling the image corresponding to the target line according to the preset polygonal block further includes:

[0020] Convert the preset polygonal block into an XML file or JSON data to obtain structured annotation information;

[0021] The structured annotation information is synchronized to the cloud and stored in association with the image corresponding to the target route.

[0022] A second aspect of the present application provides a transmission line image annotation device, comprising:

[0023] An image acquisition module, used to acquire images of a target line from multiple shooting angles;

[0024] An image classification module, used for classifying the multi-shooting angle images according to EXIF ​​information extracted from the multi-shooting angle images to obtain classified route images, wherein the EXIF ​​information includes shooting posture and sensitivity;

[0025] A feature extraction module, used to perform a texture feature extraction operation on the classified line image using a preset LBP algorithm to obtain a texture feature image;

[0026] A layer construction module, used to construct a labeled annotation layer according to the multi-shooting angle image and the texture feature image;

[0027] A block generation module, used to generate a preset polygonal block based on the annotation layer, wherein the preset polygonal block includes an angle label and a texture label;

[0028] The image annotation module is used to annotate the image corresponding to the target line according to the preset polygonal block.

[0029] Preferably, the layer construction module is specifically used for:

[0030] Constructing an initial annotation layer according to the multi-shooting angle images and the texture feature images;

[0031] Angle labels and texture labels are added to the angle layer and texture feature layer in the initial annotation layer respectively to obtain an annotation layer.

[0032] Preferably, it also includes:

[0033] A grayscale module, used for grayscale processing the multi-shooting angle images to obtain grayscale images;

[0034] Accordingly, the layer construction module is specifically used for:

[0035] A labeled annotation layer is constructed according to the multi-shooting angle images, the texture feature images and the grayscale images.

[0036] Preferably, it also includes:

[0037] A format conversion module, used to convert the preset polygonal block into an XML file or JSON data to obtain structured annotation information;

[0038] The associated storage module is used to synchronize the structured annotation information to the cloud and associate it with the image corresponding to the target route for storage.

[0039] A third aspect of the present application provides a power transmission line image annotation device, the device comprising a processor and a memory;

[0040] The memory is used to store program code and transmit the program code to the processor;

[0041] The processor is used to execute the power transmission line image annotation method described in the first aspect according to the instructions in the program code.

[0042] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium is used to store program codes, and the program codes are used to execute the power transmission line image annotation method described in the first aspect.

[0043] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0044] In the present application, a method for annotating a power transmission line image is provided, comprising: acquiring multi-shooting angle images of a target line; performing shooting classification on the multi-shooting angle images according to EXIF ​​information extracted from the multi-shooting angle images to obtain a classified line image, wherein the EXIF ​​information includes shooting posture and sensitivity; performing a texture feature extraction operation on the classified line image using a preset LBP algorithm to obtain a texture feature image; constructing a labeled annotation layer according to the multi-shooting angle images and the texture feature image; generating a preset polygonal block based on the annotation layer, wherein the preset polygonal block includes an angle label and a texture label; and annotating an image corresponding to the target line according to the preset polygonal block.

[0045] The power transmission line image annotation method provided by the present application provides a variety of image information to annotate the image of the target line, namely, shooting angle, texture features, labeling and polygonal blocks, etc., and integrates these image information to express the image of the target line, so that the differences between the objects in the image are greater, which is conducive to improving the accuracy and speed of distinguishing annotations; automatic annotation is performed based on these layer information, without the need for excessive intervention by operators, reducing dependence on human operations, and with higher efficiency and accuracy. Therefore, the present application can solve the technical problem that the annotation operation of existing image data is time-consuming and inefficient, resulting in poor data management results. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A schematic diagram of a flow chart of a method for annotating a power transmission line image provided in an embodiment of the present application;

[0047] Figure 2 A schematic diagram of the structure of a power transmission line image annotation device provided in an embodiment of the present application;

[0048] Figure 3 A schematic diagram of the application interface of the method of the present application in the identification of distribution line faults provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0050] For easier understanding, see Figure 1 , an embodiment of a power transmission line image annotation method provided by the present application includes:

[0051] Step 101: Acquire multi-angle images of a target route.

[0052] Generally speaking, when taking photos of the same target route or the same target object on the target route, the drone switches angles during the inspection process and takes multiple images and transmits them back. The images carry some shooting annotation information. Therefore, the acquired image collection is a comprehensive file that can index and construct the captured images or information to facilitate the management of image data.

[0053] Step 102: Classify the multi-shooting angle images according to the EXIF ​​information extracted from the multi-shooting angle images to obtain classified route images, wherein the EXIF ​​information includes shooting posture and sensitivity.

[0054] EXIF stands for Exchangeable image file format, which is specially designed for digital camera photos and can record the attribute information and shooting data of digital photos. It can be attached to JPEG, TIFF, RIFF and other files to add the content of digital camera shooting information and index map or version information of image processing software.

[0055] Specifically, EXIF ​​can include information such as image direction, image resolution, resolution unit, exposure time, aperture value, shooting mode, ISO sensitivity, shooting time, image size, and physical focal length of the lens; while shooting posture mainly refers to information such as image direction and focal length. Classification based on EXIF ​​information is essentially a process of screening and distinguishing images based on shooting conditions; the classification here facilitates image management.

[0056] Step 103: Use a preset LBP algorithm to perform a texture feature extraction operation on the classified line image to obtain a texture feature image.

[0057] The LBP (Local Binary Pattern) algorithm is an operator used to describe the local texture features of an image. It has significant advantages such as rotation invariance and grayscale invariance. Each LBP operator can generate different binary patterns. The original LBP operator is defined as comparing the grayscale values ​​of the 8 adjacent pixels with the center pixel of the window in a 3*3 window, with the center pixel of the window as the threshold. If the surrounding pixel value is greater than the center pixel value, the position of the pixel is marked as 1, otherwise it is 0. In this way, the 8 points in the 3*3 neighborhood can generate an 8-bit binary number (usually converted to a decimal number, namely the LBP code, a total of 256 types) after comparison, that is, the LBP value of the center pixel of the window is obtained, and this value is used to reflect the texture information of the area. Some other improved LBP algorithms may have different binary patterns, but it is obvious that the LBP operator can obtain an LBP "code" at each pixel. Then, after extracting the LBP operator of an image (which records the grayscale value of each pixel), the obtained LBP feature is still "a picture".

[0058] This embodiment samples the LBP operator to process each pixel in the image and obtains the calculated LBP value to express the texture feature corresponding to the image. The extraction of texture features can improve the boundary distinction in the image and help to mark the boundary segmentation.

[0059] Step 104: construct a labeled annotation layer based on the multi-shooting angle images and the texture feature images.

[0060] Furthermore, step 104 includes:

[0061] Construct an initial annotation layer based on multi-shooting angle images and texture feature images;

[0062] Angle labels and texture labels are added to the angle layer and texture feature layer in the initial annotation layer respectively to obtain the annotation layer.

[0063] The various layers in the annotation layer are in a superimposed positional relationship, and each layer represents different image information, such as angle and texture. Corresponding annotation information is added to different layers to achieve annotation of different levels of data.

[0064] Furthermore, step 104, before that, also includes:

[0065] Grayscale processing is performed on the images shot at multiple shooting angles to obtain grayscale images;

[0066] Accordingly, a labeled annotation layer is constructed based on the multi-angle images and texture feature images, including:

[0067] Construct a labeled annotation layer based on multi-angle images, texture feature images and grayscale images.

[0068] There is no limit on the number of specific layers in the annotation layer. An image can be described from different angles according to different needs, and then the layer is added to the annotation layer; it is understandable that the added layer also needs to add corresponding annotation information.

[0069] Step 105: Generate a preset polygonal block based on the annotation layer, where the preset polygonal block includes an angle label and a texture label.

[0070] The preset polygonal block is actually the shape of the target object in the image. No matter what device or component it is, it can be expressed by an approximate polygonal block. The polygonal block in this embodiment contains multiple layers and corresponding labels. Therefore, in addition to the polygonal block being able to describe a target object, the target object can also be described through accurate layer annotation information.

[0071] Step 106: annotate the image corresponding to the target line according to the preset polygonal block.

[0072] If the method in the embodiment of the present application is applied to a specific distribution line fault identification process, the imported target line image interface includes multiple different layers and also includes fault identification labels set according to multiple layers and polygonal block layers, such as device damage, external surface contamination, foreign matter or structural damage, etc.; please refer to Figure 3 , and can be specifically designed according to the layers and actual needs. In short, the labeling method provided in the embodiment of the present application can help operators reduce the pressure of data processing and management, and facilitate improving work efficiency.

[0073] Furthermore, step 106 further includes:

[0074] Convert the preset polygonal blocks into XML files or JSON data to obtain structured annotation information;

[0075] The structured annotation information is synchronized to the cloud and stored in association with the image corresponding to the target route.

[0076] The process of structural conversion between annotation information and images, that is, the process of associating annotation information with images and storing them. The stored information is not only images, but also annotation data or information in other formats, so the format needs to be adjusted for unified storage.

[0077] The power transmission line image annotation method provided in the embodiment of the present application provides a variety of image information to annotate the image of the target line, namely, shooting angle, texture features, labeling and polygonal blocks, etc., and integrates these image information to express the image of the target line, so that the differences between the objects in the image are greater, which is conducive to improving the accuracy and speed of distinguishing annotations; automatic annotation is performed based on these layer information, and there is no need for excessive intervention by operators, which reduces the dependence on human operations and has higher efficiency and accuracy. Therefore, the embodiment of the present application can solve the technical problem that the annotation operation of existing image data is time-consuming and inefficient, resulting in poor data management results.

[0078] For easier understanding, see Figure 2 The present application provides an embodiment of a power transmission line image annotation device, comprising:

[0079] An image acquisition module 201 is used to acquire images of a target route from multiple shooting angles;

[0080] An image classification module 202 is used to classify the multi-shooting angle images according to the EXIF ​​information extracted from the multi-shooting angle images to obtain classified route images, wherein the EXIF ​​information includes shooting posture and sensitivity;

[0081] The feature extraction module 203 is used to perform a texture feature extraction operation on the classified line image using a preset LBP algorithm to obtain a texture feature image;

[0082] A layer construction module 204 is used to construct a labeled annotation layer according to the multi-shooting angle images and the texture feature images;

[0083] The block generation module 205 is used to generate a preset polygonal block based on the annotation layer, and the preset polygonal block includes an angle label and a texture label;

[0084] The image annotation module 206 is used to annotate the image corresponding to the target line according to the preset polygonal blocks.

[0085] Furthermore, the layer construction module 204 is specifically used for:

[0086] Construct an initial annotation layer based on multi-shooting angle images and texture feature images;

[0087] Angle labels and texture labels are added to the angle layer and texture feature layer in the initial annotation layer respectively to obtain the annotation layer.

[0088] Furthermore, it also includes:

[0089] A grayscale module 207 is used to perform grayscale processing on the multi-shooting angle images to obtain grayscale images;

[0090] Accordingly, the layer construction module 204 is specifically used for:

[0091] Construct a labeled annotation layer based on multi-angle images, texture feature images and grayscale images.

[0092] Furthermore, it also includes:

[0093] A format conversion module 208 is used to convert the preset polygonal block into an XML file or JSON data to obtain structured annotation information;

[0094] The associated storage module 209 is used to synchronize the structured annotation information to the cloud and associate it with the image corresponding to the target route for storage.

[0095] For ease of understanding, the present application also provides a power transmission line image annotation device, the device comprising a processor and a memory;

[0096] The memory is used to store the program code and transmit the program code to the processor;

[0097] The processor is used to execute the power transmission line image annotation method in the above method embodiment according to the instructions in the program code.

[0098] For ease of understanding, the present application also provides a computer-readable storage medium, which is used to store program codes, and the program codes are used to execute the power transmission line image annotation method in the above method embodiment.

[0099] In the several embodiments provided in the present application, it should be understood that the disclosed 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 through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0100] 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.

[0101] In addition, each functional unit in each embodiment of the present application 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.

[0102] 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 application 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, and the computer software product is stored in a storage medium, including a number of instructions for executing all or part of the steps of the method described in each embodiment of the present application through a computer device (which can be a personal computer, a server, or a network device, etc.). The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (full name in English: Read-Only Memory, English abbreviation: ROM), random access memory (full name in English: Random Access Memory, English abbreviation: RAM), disk or optical disk and other media that can store program codes.

[0103] 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 annotating a power transmission line image, characterized in that: include: Acquire multi-angle images of the target line; Classify the multi-shooting angle images according to the EXIF ​​information extracted from the multi-shooting angle images to obtain classified line images, wherein the EXIF ​​information includes shooting posture and sensitivity; Using a preset LBP algorithm to perform a texture feature extraction operation on the classified line image to obtain a texture feature image; Constructing a labeled annotation layer according to the multi-shooting angle images and the texture feature images; Generate a preset polygonal block based on the annotation layer, wherein the preset polygonal block includes an angle label and a texture label; The image corresponding to the target line is labeled according to the preset polygonal block.

2. The power transmission line image annotation method according to claim 1, characterized in that: The step of constructing a labeled annotation layer according to the multi-shooting angle image and the texture feature image includes: Constructing an initial annotation layer according to the multi-shooting angle images and the texture feature images; Angle labels and texture labels are added to the angle layer and texture feature layer in the initial annotation layer respectively to obtain an annotation layer.

3. The power transmission line image annotation method according to claim 1, characterized in that: The step of constructing a labeled annotation layer according to the multi-shooting angle image and the texture feature image also includes: Performing grayscale processing on the multi-shooting angle images to obtain grayscale images; Accordingly, constructing a labeled annotation layer according to the multi-shooting angle image and the texture feature image includes: A labeled annotation layer is constructed according to the multi-shooting angle images, the texture feature images and the grayscale images.

4. The power transmission line image annotation method according to claim 1, characterized in that: The process of labeling the image corresponding to the target line according to the preset polygonal block further includes: Convert the preset polygonal block into an XML file or JSON data to obtain structured annotation information; The structured annotation information is synchronized to the cloud and stored in association with the image corresponding to the target route.

5. A transmission line image annotation device, characterized in that: include: An image acquisition module, used to acquire images of a target line from multiple shooting angles; An image classification module, used for classifying the multi-shooting angle images according to EXIF ​​information extracted from the multi-shooting angle images to obtain classified route images, wherein the EXIF ​​information includes shooting posture and sensitivity; A feature extraction module, used to perform a texture feature extraction operation on the classified line image using a preset LBP algorithm to obtain a texture feature image; A layer construction module, used to construct a labeled annotation layer according to the multi-shooting angle image and the texture feature image; A block generation module, used to generate a preset polygonal block based on the annotation layer, wherein the preset polygonal block includes an angle label and a texture label; The image annotation module is used to annotate the image corresponding to the target line according to the preset polygonal block.

6. The power transmission line image annotation device according to claim 5, characterized in that: The layer construction module is specifically used for: Constructing an initial annotation layer according to the multi-shooting angle images and the texture feature images; Angle labels and texture labels are added to the angle layer and texture feature layer in the initial annotation layer respectively to obtain an annotation layer.

7. The power transmission line image annotation device according to claim 5, characterized in that: Also includes: A grayscale module, used for grayscale processing the multi-shooting angle images to obtain grayscale images; Accordingly, the layer construction module is specifically used for: A labeled annotation layer is constructed according to the multi-shooting angle images, the texture feature images and the grayscale images.

8. The power transmission line image annotation device according to claim 5, characterized in that: Also includes: A format conversion module, used to convert the preset polygonal block into an XML file or JSON data to obtain structured annotation information; The associated storage module is used to synchronize the structured annotation information to the cloud and associate it with the image corresponding to the target route for storage.

9. A transmission line image annotation device, characterized in that: The device comprises a processor and a memory; The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the power transmission line image annotation method according to any one of claims 1 to 4 according to the instructions in the program code.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store program codes, and the program codes are used to execute the power transmission line image annotation method according to any one of claims 1 to 4.

Images