Power transmission line forest fire smoke detection method and system

Abstract

The invention discloses a power transmission line mountain fire smoke detection method and system, and the method comprises the steps: segmenting a mountain fire smoke candidate region in a to-be-detected power transmission channel picture through a pre-trained semantic segmentation model, and calculating a semantic segmentation confidence score; for each mountain fire smoke candidate area, calculating a shape similarity score through a shape matching method, calculating a position confidence score through a position function, and calculating a color confidence score through color distribution comparison; and carrying out weighted summation on the semantic segmentation confidence score, the shape similarity score, the position confidence score and the color confidence score to obtain a final smoke confidence score, and comparing the final smoke confidence score with a preset threshold to judge whether the corresponding mountain fire smoke candidate region is mountain fire smoke. According to the technical scheme provided by the invention, the forest fire smoke condition in the power transmission line channel picture can be effectively identified, the misidentification conditions of pond smoke, chimney smoke, cloud and mist and the like are obviously reduced, and the method has a wide application scene.

Description

Technical Field

[0001] This invention relates to the field of image recognition, specifically to a method and system for detecting smoke from wildfires on power transmission lines. Background Technology

[0002] With the rapid development of my country's power industry and the continuous expansion of the power grid, transmission lines, as the main arteries for power transmission, are crucial for safe and stable operation. However, transmission lines often traverse complex geographical environments such as mountains and forests, facing various natural risks. Among these, wildfires, due to their sudden onset and destructive power, have become one of the main factors endangering the safety of transmission lines. The dense smoke produced by wildfires can cause flashovers in insulator strings, leading to line tripping and widespread power outages, posing a serious challenge to power grid stability and socio-economic activities. Therefore, timely and accurate monitoring of wildfire smoke along transmission line corridors is a vital link in the power system's disaster prevention and mitigation efforts.

[0003] Currently, methods for detecting smoke from wildfires along power transmission lines have many shortcomings. For example, traditional threshold-based detection methods are poorly adaptable to complex environments. Fixed thresholds are insufficient to handle the ever-changing natural conditions. High temperatures in summer can lead to widespread increases in ambient temperature, easily causing false alarms. Sensors are also susceptible to dust and rain contamination, leading to decreased sensitivity and missed detections. Furthermore, single-point sensors can only cover a limited area, requiring the deployment of numerous devices for full-line monitoring, resulting in high costs and difficult maintenance. This is especially problematic since wildfires often originate in remote locations, making them difficult to capture effectively by sparsely distributed sensor networks. While video surveillance has become widespread, image analysis using cameras has been applied. However, relying solely on image recognition is easily affected by environmental interference such as clouds, fog, moisture, and shadows, resulting in low accuracy. Therefore, there is an urgent need for a method and system that can accurately and reliably detect smoke from wildfires along power transmission lines. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method for detecting wildfire smoke on power transmission lines, which can more accurately identify wildfire disasters on power transmission lines, while also taking into account low cost and is suitable for widespread application.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention includes the following steps: S1, For the image of the power transmission channel to be detected, the pre-trained semantic segmentation model is used to segment the candidate region of wildfire smoke in the image and calculate the semantic segmentation confidence score Score1. S2, For each candidate wildfire smoke region, calculate the shape similarity score Score2 by shape matching, the position confidence score Score3 by position function, and the color confidence score Score4 by color distribution comparison. S3, the semantic segmentation confidence score Score1, shape similarity score Score2, location confidence score Score3 and color confidence score Score4 are weighted and summed to obtain the final smoke confidence score Score. The final smoke confidence score Score is compared with a preset threshold to determine whether the corresponding wildfire smoke candidate region is wildfire smoke. Furthermore, in step S1, when segmenting candidate regions of wildfire smoke in the image using the pre-trained semantic segmentation model and calculating the semantic segmentation confidence score Score1, the function expression for calculating the semantic segmentation confidence score Score1 is as follows: ; in, Indicates candidate areas for wildfire smoke. This represents the pixels within the candidate area of ​​wildfire smoke. Represents pixels The probability that it belongs to the category of wildfire smoke. Indicates candidate areas for wildfire smoke The total number of pixels.

[0006] Furthermore, in step S2, calculating the shape similarity score Score2 using the shape matching method includes: S2.1, the wildfire smoke candidate region is divided into multiple sub-regions around the region center according to a specified angle. For each sub-region i, a direction vector is constructed based on its two edge vertices. This yields the direction vectors of all sub-regions. The vector features constituted ; S2.2, vector features And the vector features of each wildfire smoke region in the given wildfire smoke region vector feature library. Calculate the similarity S: ; The wildfire smoke region vector feature library includes vector features of various known wildfire smoke regions. ; S2.3, select the largest similarity S as the shape similarity score Score2 for the candidate wildfire smoke region.

[0007] Furthermore, the calculation function expression for the location confidence score Score3 in step S2 is as follows: ; in, This indicates the height of the image of the power transmission channel to be inspected. This indicates the uppermost point of the candidate area for wildfire smoke. This represents the lowest point of the candidate wildfire smoke area, and satisfies... .

[0008] Furthermore, the calculation function expression for the color confidence score Score4 in step S2 is as follows: ; in, This represents the variance of the RGB three-channel color values ​​of the candidate region for wildfire smoke. This represents the average of the RGB three-channel color values ​​of the candidate region for wildfire smoke. This represents the average of the RGB three-channel color values ​​of the candidate region for wildfire smoke.

[0009] Furthermore, the formula for calculating the final smoke confidence score in step S3 is as follows: ; in, , , , The weighted scores for semantic segmentation confidence score Score1, shape similarity score Score2, position confidence score Score3, and color confidence score Score4 are respectively, and satisfy the following: .

[0010] Furthermore, the loss function used during training of the pre-trained semantic segmentation model in step S1 is expressed as follows: ; in, Represents the loss function. Indicates the number of samples. and These represent the height and width of the power transmission channel image to be inspected, respectively. This indicates the total number of categories, which includes at least smoke categories and non-smoke categories. Indicates sample median coordinate The pixel at that location belongs to the category Indicator variable, when sample median coordinate The pixel at that location belongs to the category The value is 1 if it is true, and 0 otherwise. This represents the samples predicted by the semantic segmentation model. median coordinate The pixel at that location belongs to the category The probability of.

[0011] The present invention also proposes a wildfire smoke detection system, comprising an interconnected microprocessor and a memory, wherein the microprocessor is programmed or configured to execute the above-described method for detecting wildfire smoke on transmission lines.

[0012] The present invention also proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0013] The present invention also proposes a computer program product, comprising a computer program that, when executed by a processor, implements the steps of the above-described method.

[0014] Compared with the prior art, the advantages of the present invention are as follows: This invention, by comprehensively considering semantic segmentation confidence score, shape similarity score, location confidence score, and color confidence score, can effectively identify wildfire smoke in images of power transmission line corridors and significantly reduce misidentification of smoke such as pond smoke, chimney smoke, and clouds, and has a wide range of applications. Attached Figure Description

[0015] Figure 1 The flowchart of this embodiment. Detailed Implementation

[0016] The present invention will be further described below with reference to the accompanying drawings and specific preferred embodiments, but this does not limit the scope of protection of the present invention.

[0017] The technical solution adopted in this embodiment is as follows: Figure 1 As shown, it includes the following steps: S1, For the image of the power transmission channel to be detected, the pre-trained semantic segmentation model is used to segment the candidate region of wildfire smoke in the image and calculate the semantic segmentation confidence score Score1. S2, For each candidate wildfire smoke region, calculate the shape similarity score Score2 by shape matching, the position confidence score Score3 by position function, and the color confidence score Score4 by color distribution comparison. S3, the semantic segmentation confidence score Score1, shape similarity score Score2, location confidence score Score3 and color confidence score Score4 are weighted and summed to obtain the final smoke confidence score Score. The final smoke confidence score Score is compared with a preset threshold to determine whether the corresponding wildfire smoke candidate region is wildfire smoke. This embodiment, by comprehensively considering semantic segmentation confidence score, shape similarity score, location confidence score, and color confidence score, can effectively identify wildfire smoke in images of power transmission line corridors and significantly reduce misidentification of smoke such as pond smoke, chimney smoke, and clouds, and has a wide range of application scenarios.

[0018] Preferably, in step S1, when segmenting the candidate regions of wildfire smoke in the image using the pre-trained semantic segmentation model and calculating the semantic segmentation confidence score Score1, the function expression for calculating the semantic segmentation confidence score Score1 is as follows: ; in, Indicates candidate areas for wildfire smoke. This represents the pixels within the candidate area of ​​wildfire smoke. Represents pixels The probability that it belongs to the category of wildfire smoke. Indicates candidate areas for wildfire smoke The total number of pixels.

[0019] Preferably, in step S2, calculating the shape similarity score Score2 using the shape matching method includes: S2.1, the wildfire smoke candidate region is divided into multiple sub-regions around the region center according to a specified angle. For each sub-region i, a direction vector is constructed based on its two edge vertices. This yields the direction vectors of all sub-regions. The vector features constituted ; S2.2, vector features And the vector features of each wildfire smoke region in the given wildfire smoke region vector feature library. Calculate the similarity S: ; The wildfire smoke region vector feature library includes vector features of various known wildfire smoke regions. ; S2.3, select the largest similarity S as the shape similarity score Score2 for the candidate wildfire smoke region.

[0020] In this embodiment, for each candidate wildfire smoke region, the region edge is divided into 360 equal parts around the region center, and a direction vector can be obtained for each pair of edge vertices. Starting from the central perpendicular vertex of the region, a list of direction vectors is obtained by proceeding clockwise. , ,..., This yields a vector feature V1 containing 360 elements; the vector feature of the known wildfire smoke area. The vector features are generated for images in the image dataset of labeled wildfire areas using the vector feature V1 generation method described above.

[0021] Preferably, the calculation function expression for the location confidence score Score3 in step S2 is: ; in, This indicates the height of the image of the power transmission channel to be inspected. This indicates the uppermost point of the candidate area for wildfire smoke. This represents the lowest point of the candidate wildfire smoke area, and satisfies... .

[0022] The dividing line is defined as the point at half the height of the power transmission channel image. For candidate wildfire smoke areas, if the entire area is below the dividing line, the location confidence score is 1; if the dividing line passes through the area, the location confidence score is the proportion of the dividing line within the height of the candidate wildfire smoke area from bottom to top; if the entire area is above the dividing line, the location confidence score is 0. Preferably, the calculation function expression for the color confidence score Score4 in step S2 is: ; in, This represents the variance of the RGB three-channel color values ​​of the candidate region for wildfire smoke. This represents the average of the RGB three-channel color values ​​of the candidate region for wildfire smoke. This represents the average of the RGB three-channel color values ​​of the candidate region for wildfire smoke.

[0023] Preferably, the formula for calculating the final smoke confidence score in step S3 is as follows: ; in, , , , The weighted scores for semantic segmentation confidence score Score1, shape similarity score Score2, position confidence score Score3, and color confidence score Score4 are respectively, and satisfy the following: .

[0024] Preferably, the loss function used during training of the pre-trained semantic segmentation model in step S1 has the following expression: ; in, Represents the loss function. Indicates the number of samples. and These represent the height and width of the power transmission channel image to be inspected, respectively. This indicates the total number of categories, which includes at least smoke categories and non-smoke categories. Indicates sample median coordinate The pixel at that location belongs to the category Indicator variable, when sample median coordinate The pixel at that location belongs to the category The value is 1 if it is true, and 0 otherwise. This represents the samples predicted by the semantic segmentation model. median coordinate The pixel at that location belongs to the category The probability of.

[0025] This embodiment also proposes a wildfire smoke detection system, including an interconnected microprocessor and a memory, wherein the microprocessor is programmed or configured to execute the above-described method for detecting wildfire smoke on power transmission lines.

[0026] This embodiment also proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0027] This embodiment also proposes a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method.

[0028] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0029] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for detecting smoke from wildfires on power transmission lines, characterized in that, Includes the following steps: S1, For the image of the power transmission channel to be detected, the pre-trained semantic segmentation model is used to segment the candidate region of wildfire smoke in the image and calculate the semantic segmentation confidence score Score1. S2, For each candidate wildfire smoke region, calculate the shape similarity score Score2 by shape matching, the position confidence score Score3 by position function, and the color confidence score Score4 by color distribution comparison. S3, the semantic segmentation confidence score Score1, shape similarity score Score2, location confidence score Score3 and color confidence score Score4 are weighted and summed to obtain the final smoke confidence score Score. The final smoke confidence score Score is compared with a preset threshold to determine whether the corresponding wildfire smoke candidate region is wildfire smoke.

2. The method for detecting wildfire smoke on transmission lines according to claim 1, characterized in that, In step S1, when using the pre-trained semantic segmentation model to segment candidate regions of wildfire smoke in the image and calculating the semantic segmentation confidence score Score1, the function expression for calculating the semantic segmentation confidence score Score1 is as follows: ； in, Indicates candidate areas for wildfire smoke. This represents the pixels within the candidate area of ​​wildfire smoke. Represents pixels The probability that it belongs to the category of wildfire smoke. Indicates candidate areas for wildfire smoke The total number of pixels.

3. The method for detecting wildfire smoke on transmission lines according to claim 1, characterized in that, In step S2, calculating the shape similarity score Score2 using the shape matching method includes: S2.1, the wildfire smoke candidate region is divided into multiple sub-regions around the region center according to a specified angle. For each sub-region i, a direction vector is constructed based on its two edge vertices. This yields the direction vectors of all sub-regions. The vector features constituted ; S2.2, vector features And the vector features of each wildfire smoke region in the given wildfire smoke region vector feature library. Calculate the similarity S: ； The wildfire smoke region vector feature library includes vector features of various known wildfire smoke regions. ; S2.3, select the largest similarity S as the shape similarity score Score2 for the candidate wildfire smoke region.

4. The method for detecting wildfire smoke on transmission lines according to claim 1, characterized in that, The function expression for calculating the location confidence score Score3 in step S2 is: ； in, This indicates the height of the image of the power transmission channel to be inspected. This indicates the uppermost point of the candidate area for wildfire smoke. This represents the lowest point of the candidate wildfire smoke area, and satisfies... .

5. The method for detecting wildfire smoke on transmission lines according to claim 1, characterized in that, The function expression for calculating the color confidence score Score4 in step S2 is: ； in, This represents the variance of the RGB three-channel color values ​​of the candidate region for wildfire smoke. This represents the average of the RGB three-channel color values ​​of the candidate region for wildfire smoke. This represents the average of the RGB three-channel color values ​​of the candidate region for wildfire smoke.

6. The method for detecting wildfire smoke on transmission lines according to claim 1, characterized in that, The formula for calculating the final smoke confidence score in step S3 is as follows: ； in, , , , The weighted scores for semantic segmentation confidence score Score1, shape similarity score Score2, position confidence score Score3, and color confidence score Score4 are respectively, and satisfy the following: 。 7. The method for detecting wildfire smoke on transmission lines according to claim 1, characterized in that, The loss function used during training of the pre-trained semantic segmentation model in step S1 is expressed as follows: ； in, Represents the loss function. Indicates the number of samples. and These represent the height and width of the power transmission channel image to be inspected, respectively. This indicates the total number of categories, which includes at least smoke categories and non-smoke categories. Indicates sample median coordinate The pixel at that location belongs to the category Indicator variable, when sample median coordinate The pixel at that location belongs to the category The value is 1 if it is true, and 0 otherwise. This represents the samples predicted by the semantic segmentation model. median coordinate The pixel at that location belongs to the category The probability of.

8. A power transmission line wildfire smoke detection system, comprising a microprocessor and a memory interconnected, characterized in that, The microprocessor is programmed or configured to perform the transmission line wildfire smoke detection method according to any one of claims 1 to 7.

9. A computer-readable storage medium storing a computer program or instructions, characterized in that, The computer program or instructions are programmed or configured to execute, via a processor, the method for detecting wildfire smoke along transmission lines as described in any one of claims 1 to 7.

10. A computer program product, comprising a computer program or instructions, characterized in that, The computer program or instructions are programmed or configured to execute, via a processor, the method for detecting wildfire smoke along transmission lines as described in any one of claims 1 to 7.

Images