A method and system for quantitative recognition of transmission line monitoring images

Abstract

The present application relates to the field of image processing technology, and specifically provides a method and system for quantitative recognition of transmission line monitoring images, the method comprising: obtaining monitoring images and equipment parameters; recognizing the monitoring images according to the convolution algorithm to generate recognition images; inputting the recognition images into a monocular depth estimation model to generate depth information; geometrically analyzing the depth information to generate relative height information; obtaining a depth threshold from the depth information according to the transmission line tower spacing and safety distance; generating a height range according to the line elevation value, sag limit, safety distance and relative height information; generating a mask image according to the depth threshold and the height range; performing an AND operation on the recognition image and the mask image to generate a recognition result; and pushing the recognition image to a display terminal according to the recognition result. By performing an AND operation on the recognition image and the mask image, it is possible to directly determine whether the current monitoring image constitutes a safety hazard, thereby avoiding the occurrence of false alarms.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method and system for quantitatively identifying transmission line monitoring images. Background Art

[0002] Image recognition has been applied to the intelligent transformation of power transmission line monitoring. The data collected by the monitoring device is simply identified and then returned to the background, where the background staff will identify and deal with the identification situation. At present, the monitoring device is at the front end of management and control, facing problems with power consumption and data transmission. It is relatively simple to identify the target. At the same time, factors such as the large range of transmission line scenes and changes in lighting cause many false alarms in the identification results returned to the background. The identification results need to be screened by humans in shifts all day at the display terminal to determine whether the identification is correct or not, which is time-consuming and labor-intensive.

[0003] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0004] To this end, the technical problem to be solved by the present invention is to overcome the problem that target identification in the prior art is relatively simple. At the same time, factors such as the large range of transmission line scenes and changes in lighting result many false alarms in the identification results returned to the background. In addition, the identification results require manual shifts all day to screen the display terminal to see if they are correct, which is time-consuming and labor-intensive.

[0005] In order to solve the above technical problems, the first aspect of the present invention provides a transmission line monitoring image quantitative recognition method, the method comprising: acquiring a monitoring image and equipment parameters; the monitoring image comprises a target depth value; the equipment parameters comprise: transmission line tower spacing, safety distance, line elevation value and sag limit value;

[0006] Recognize the monitoring image according to a convolution algorithm to generate a recognition image;

[0007] Inputting the recognition image into a monocular depth estimation model to generate depth information;

[0008] Performing geometric analysis on the depth information to generate relative height information;

[0009] Obtaining a depth threshold from the depth information according to the transmission line tower spacing and the safety distance;

[0010] Generate a height range according to the line elevation value, the sag limit value, the safety distance and the relative height information;

[0011] generating a mask image according to the depth threshold and the height range;

[0012] Performing an AND operation on the recognition image and the mask image to generate a recognition result;

[0013] The recognized image is pushed to a display terminal according to the recognition result.

[0014] In one embodiment of the present invention, the steps before inputting the recognition image into the monocular depth estimation model include:

[0015] The monitoring image and the target depth value are input into a deep neural network to generate a monocular depth estimation model.

[0016] In one embodiment of the present invention, the step of generating a mask image according to the depth threshold and the height range comprises:

[0017] forming a focus area by using the depth threshold and the height range envelope;

[0018] A mask image is generated according to the focused area.

[0019] In one embodiment of the present invention, the step of performing an AND operation on the recognition image and the mask image to generate a recognition result includes:

[0020] Performing an AND operation on the recognition image and the mask image to generate an operation result;

[0021] Generate the recognition result according to the operation result;

[0022] If the operation result is false, the recognition result does not constitute a safety hazard, and the recognition image is stored;

[0023] If the operation result is true, the recognition result constitutes a safety hazard, and the recognition image is pushed to the display terminal.

[0024] In one embodiment of the present invention, the step of performing an AND operation on the recognition image and the mask image to generate an operation result includes:

[0025] Determining whether the recognition image is within the mask image;

[0026] If the recognition image is within the mask image, the operation result is true;

[0027] If the identification image is not within the mask image, the operation result is false.

[0028] In one embodiment of the present invention, the step of performing an AND operation on the recognition image and the mask image to generate an operation result further includes:

[0029] determining whether the recognition image is partially within the mask image;

[0030] If the identified image is partially within the mask image, the operation result is true;

[0031] If no part of the identification image is within the mask image, the operation result is false.

[0032] In one embodiment of the present invention, the steps after storing the recognition image include:

[0033] The identified image is named according to a preset rule.

[0034] A second aspect of the present invention provides a transmission line monitoring image quantification recognition system, which is applied to a method proposed in any one of the first aspects above, the system comprising: a data acquisition module, a recognition module, a calculation module and an operation module;

[0035] The data acquisition module is configured to: acquire monitoring images and equipment parameters; the monitoring images include target depth values; the equipment parameters include: transmission line tower spacing, safety distance, line elevation value, sag limit value and safety distance limit value;

[0036] The recognition module is configured to: recognize the monitoring image according to the convolution algorithm to generate a recognition image; input the recognition image into a monocular depth estimation model to generate depth information; perform geometric analysis on the depth information to generate relative height information;

[0037] The calculation module is configured to: obtain a depth threshold from the depth information according to the transmission line tower spacing and the safety distance; obtain a height range from the relative height information according to the line elevation value, the sag limit value and the safety distance limit value; generate a mask image according to the depth threshold and the height range;

[0038] The operation module is configured to: perform an AND operation on the recognition image and the mask image to generate a recognition result; and push the recognition image to a display terminal according to the recognition result.

[0039] The third aspect of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method described in the first aspect or any possible implementation of the first aspect is implemented.

[0040] A fourth aspect of the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect or any possible implementation of the first aspect.

[0041] The above technical solution of the present invention has the following advantages compared with the prior art:

[0042] The invention discloses a method and system for quantitatively identifying a power transmission line monitoring image. By performing AND operations on the identification image and the mask image, it is possible to directly determine whether the current monitoring image constitutes a potential safety hazard, thereby avoiding false alarms and further improving screening efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to make the contents of the present invention more clearly understood, the present invention is further described in detail below based on specific embodiments of the present invention in conjunction with the accompanying drawings.

[0044] Figure 1 It is a flow chart of a transmission line monitoring image quantitative recognition method and system provided by the present invention;

[0045] Figure 2 A system architecture diagram of a transmission line monitoring image quantization recognition method and system provided by the present invention. DETAILED DESCRIPTION

[0046] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0047] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0048] Reference Figure 1 As shown, the present invention provides a method for quantitative recognition of transmission line monitoring images, the method comprising:

[0049] S100, acquiring monitoring images and equipment parameters; the monitoring images include a target depth value; the equipment parameters include: transmission line tower spacing, safety distance, line elevation value and sag limit value;

[0050] In step S100, the power transmission line monitoring device is generally fixedly installed, and the monitoring image and equipment parameters of various equipment are collected through the monitoring device, wherein the monitoring image also includes the target depth value in the image, and the equipment parameters include the transmission line tower spacing, safety distance, line elevation value, sag limit value and safety distance limit value. The line elevation value refers to the distance from a certain point along the plumb line direction to the absolute base surface, and the sag limit value is the distance between the maximum sag of the overhead line and the ground.

[0051] S200, identifying the monitoring image according to a convolution algorithm to generate a recognition image;

[0052] In step S200, the monitoring image is recognized by a convolution algorithm to obtain a recognition image, thereby realizing the recognition of common intrusions (cranes, trailers, excavators, etc.) and targets such as wildfires.

[0053] S300, inputting the recognition image into a monocular depth estimation model to generate depth information;

[0054] In step S300, the steps before inputting the recognition image into the monocular depth estimation model include: inputting the monitoring image and the target depth value into a deep neural network to generate a monocular depth estimation model.

[0055] In actual application scenarios, the monocular depth estimation model is trained based on a training data set, which includes monitoring images collected by the monitoring device and target depth values ​​included in the monitoring images. The monitoring images and target depth values ​​are input into the deep neural network to train the monocular depth estimation model. When the identification image to be screened is input again, the monocular depth estimation model can evaluate the depth information of each area in the identification image, that is, the depth map. Among them, the height information of the line tower and the intrusion is used as a constraint condition of the monocular depth estimation model. After adding this constraint condition, the results obtained by the monocular depth estimation model can be more accurate.

[0056] S400, performing geometric analysis on the depth information to generate relative height information;

[0057] In step S400, the depth information of each region is analyzed according to the analysis to obtain the relative height information of the target in the recognition image.

[0058] S500, acquiring a depth threshold from the depth information according to the transmission line tower spacing and the safety distance;

[0059] In step S500, a corresponding depth threshold is selected from the depth information based on the transmission line tower spacing and the safety distance.

[0060] S600, generating a height range according to the line elevation value, the sag limit value, the safety distance and the relative height information;

[0061] In step S600, the corresponding height range is defined according to the line elevation, sag limit, safety distance and relative height information, wherein the line elevation value refers to the distance from a certain point to the absolute base plane along the plumb line direction, and the sag limit is the distance between the maximum sag of the overhead line and the ground.

[0062] S700, generating a mask image according to the depth threshold and the height range;

[0063] In step S700, the step of generating a mask image according to the depth threshold and the height range includes: forming a focus area by enveloping the depth threshold and the height range; and generating a mask image according to the focus area.

[0064] In actual application scenarios, the area formed by the depth threshold and the height range envelope is the focus area. The mask image of the focus area is obtained, and the mask image is constructed for each fixed camera. The intrinsic and extrinsic parameters of the camera are known when installed, and the scale factor of the scene is obtained in advance based on the pinhole imaging model. The actual scene focus area can be determined according to the threshold. The actual scene is the world coordinate, and the captured image has image coordinates. The vertex coordinates of the actual scene focus area can be obtained according to the tower coordinates, and the vertex coordinates are mapped to the image coordinates to obtain the mask image.

[0065] The threshold determination method of the actual scene focus area is as follows: the coordinates of the tower are known, the camera is installed on the tower, and its coordinates are also known. There are power lines between the towers, and the line sag can be calculated according to relevant regulations. The corresponding voltage level can get the safety distance. Therefore, when the depth is within the range of the camera's visible depth to the depth of the next tower, the height greater than (maximum sag-safety distance) and the height less than (tower height + safety distance) are both the height of interest, and the width is the left and right boundaries of the tower + safety distance + 10m; the depth is within the range of the tower safety distance + 10m to the tower depth, and the height is within the range of the tower base to (tower height + safety distance), that is, an L-shaped body in space is formed, and the L-shaped body has a total of 10 vertices. According to the camera's intrinsic parameters, extrinsic parameters and the scale factor of the scene, the 10 vertices are mapped to the imaging plane, and the mask image is obtained.

[0066] S800, performing an AND operation on the recognition image and the mask image to generate a recognition result;

[0067] In step S800, the recognition image and the mask image are subjected to an AND operation, and the steps of generating a recognition result include: performing an AND operation on the recognition image and the mask image to generate an operation result; generating the recognition result according to the operation result; if the operation result is false, the recognition result does not constitute a safety hazard, and the recognition image is stored; if the operation result is true, the recognition result constitutes a safety hazard, and the recognition image is pushed to a display terminal.

[0068] The step of performing an AND operation on the recognition image and the mask image to generate an operation result includes: determining whether the recognition image is within the mask image; if the recognition image is within the mask image, the operation result is true; if the recognition image is not within the mask image, the operation result is false.

[0069] The step of performing an AND operation on the recognition image and the mask image to generate an operation result further includes: determining whether the recognition image is partially within the mask image; if the recognition image is partially within the mask image, the operation result is true; if the recognition image is not partially within the mask image, the operation result is false. The step after storing the recognition image includes: naming the recognition image according to a preset rule.

[0070] In actual application scenarios, an AND operation is performed on the recognition image and the mask image, that is, to determine whether the recognition image is within the mask image. If the recognition image is within the mask image, the result of the AND operation is true. At this time, if the recognition result corresponding to the operation result is the target, it will pose a safety hazard. In this case, the recognition image needs to be pushed to the display terminal, and the staff will evaluate, judge, and perform maintenance on the recognition image.

[0071] If the recognition image is not in the mask image, the result of the operation is false. At this time, the recognition result corresponding to the operation result is that the target will not pose a safety hazard. Then the recognition image needs to be stored, and the recognition image needs to be named according to the preset rules when storing. The preset rules are the image naming rules established by the State Grid for line inspection, which are generally clockwise, from bottom to top, and from left to right. Naming is to locate the specific equipment and components for image shooting, and it can also archive data in an orderly manner. It can also be used as sample annotation information to train image recognition algorithms, and it can also be used for historical data comparison. After that, it is necessary to wait for the next round of transmission recognition and mask operations of the monitoring image of the same scene until the operation result is true, and then push the recognition image to the display terminal.

[0072] Since the recognition image has a certain width of information, it is necessary to determine whether the recognition image is partially within the mask image. If the recognition image is partially within the mask image, then the operation result of the operation is true, and the recognition result corresponding to the operation result is the target, which will pose a safety hazard. In this case, the recognition image needs to be pushed to the display terminal, and the staff will evaluate, judge, and perform maintenance on the recognition image.

[0073] If no part of the recognition image is within the mask image, the result of the operation is false, and the recognition result corresponding to the operation result is the target, which does not pose a safety hazard. The recognition image needs to be stored, and the recognition image needs to be named according to the preset rules when storing. After that, it is necessary to wait for the next round of transmission recognition and mask operation of the monitoring image of the same scene, until the operation result is true, and then push the recognition image to the display terminal.

[0074] S900: Push the recognized image to a display terminal according to the recognition result.

[0075] In step S900, if the operation result of the operation is true, then the recognition result corresponding to the operation result will constitute a safety hazard for the target. The recognition image needs to be pushed to the display terminal based on the recognition result, and the staff will evaluate, judge and perform operation and maintenance on the recognition image.

[0076] Second, refer to Figure 2 As shown, the present application provides a transmission line monitoring image quantification recognition system, the system comprising: a data acquisition module 100, a recognition module 200, a calculation module 300 and an operation module 400;

[0077] The data acquisition module 100 is configured to: acquire monitoring images and equipment parameters; the monitoring images include target depth values; the equipment parameters include: transmission line tower spacing, safety distance, line elevation value, sag limit value and safety distance limit value;

[0078] The recognition module 200 is configured to: recognize the monitoring image according to the convolution algorithm to generate a recognition image; input the recognition image into a monocular depth estimation model to generate depth information; perform geometric analysis on the depth information to generate relative height information;

[0079] The calculation module 300 is configured to: obtain a depth threshold from the depth information according to the transmission line tower spacing and the safety distance; obtain a height range from the relative height information according to the line elevation value, the sag limit value and the safety distance limit value; generate a mask image according to the depth threshold and the height range;

[0080] The operation module 400 is configured to: perform an AND operation on the recognition image and the mask image to generate a recognition result; and push the recognition image to a display terminal according to the recognition result.

[0081] The effects of the above-mentioned system when applying the above-mentioned method can be found in the description of the above-mentioned method embodiment, which will not be repeated here.

[0082] The third aspect of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method described in the first aspect or any possible implementation of the first aspect is implemented.

[0083] A fourth aspect of the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect or any possible implementation of the first aspect.

[0084] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0085] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0086] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0087] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0088] Obviously, the above embodiments are merely examples for clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from these are still within the protection scope of the invention.

Claims

1. A method for quantitative recognition of transmission line monitoring images, characterized in that: The method comprises: Acquire monitoring images and equipment parameters; the monitoring images include target depth values; the equipment parameters include: transmission line tower spacing, safety distance, line elevation value and sag limit value; Recognize the monitoring image according to a convolution algorithm to generate a recognition image; Inputting the recognition image into a monocular depth estimation model to generate depth information; Performing geometric analysis on the depth information to generate relative height information; Obtaining a depth threshold from the depth information according to the transmission line tower spacing and the safety distance; Generate a height range according to the line elevation value, the sag limit value, the safety distance and the relative height information; generating a mask image according to the depth threshold and the height range; Performing an AND operation on the recognition image and the mask image to generate a recognition result; Pushing the recognized image to a display terminal according to the recognition result; The step of generating a mask image according to the depth threshold and the height range comprises: forming a focus area by using the depth threshold and the height range envelope; A mask image is generated according to the focused area.

2. A method for quantitative recognition of power transmission line monitoring images according to claim 1, characterized in that: The steps before inputting the recognition image into the monocular depth estimation model include: The monitoring image and the target depth value are input into a deep neural network to generate a monocular depth estimation model.

3. The method for quantitative recognition of power transmission line monitoring images according to claim 1, characterized in that: The step of performing an AND operation on the recognition image and the mask image to generate a recognition result comprises: Performing an AND operation on the recognition image and the mask image to generate an operation result; Generate the recognition result according to the operation result; If the operation result is false, the recognition result does not constitute a safety hazard, and the recognition image is stored; If the operation result is true, the recognition result constitutes a safety hazard, and the recognition image is pushed to the display terminal.

4. A method for quantitative recognition of power transmission line monitoring images according to claim 3, characterized in that: The step of performing an AND operation on the recognition image and the mask image to generate a recognition result comprises: Determining whether the recognition image is within the mask image; If the recognition image is within the mask image, the operation result is true; If the identification image is not within the mask image, the operation result is false.

5. The method for quantitative recognition of power transmission line monitoring images according to claim 1, characterized in that: The step of performing an AND operation on the recognition image and the mask image to generate a recognition result further includes: determining whether the recognition image is partially within the mask image; If the identified image is partially within the mask image, the operation result is true; If no part of the identification image is within the mask image, the operation result is false.

6. A method for quantitative recognition of power transmission line monitoring images according to claim 3, characterized in that: The steps after storing the identified image include: The identified image is named according to a preset rule.

7. A transmission line monitoring image quantification recognition system, characterized in that: A method for quantitative recognition of transmission line monitoring images applied to any one of claims 1 to 6 above, the system comprising: a data acquisition module, a recognition module, a calculation module and an operation module; The data acquisition module is configured to: acquire monitoring images and equipment parameters; the monitoring images include target depth values; the equipment parameters include: transmission line tower spacing, safety distance, line elevation value, sag limit value and safety distance limit value; The recognition module is configured to: recognize the monitoring image according to the convolution algorithm to generate a recognition image; input the recognition image into a monocular depth estimation model to generate depth information; perform geometric analysis on the depth information to generate relative height information; The calculation module is configured to: obtain a depth threshold from the depth information according to the transmission line tower spacing and the safety distance; obtain a height range from the relative height information according to the line elevation value, the sag limit value and the safety distance limit value; generate a mask image according to the depth threshold and the height range; The operation module is configured to: perform an AND operation on the recognition image and the mask image to generate a recognition result; and push the recognition image to a display terminal according to the recognition result.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method for quantitative recognition of transmission line monitoring images as described in any one of claims 1 to 6 are implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a method for quantitative recognition of transmission line monitoring images as claimed in any one of claims 1 to 6 are implemented.

Images