Transmission line hidden danger target identification method and computer readable storage medium

Through the combination of deep learning and edge detection technology, the physical distance in the transmission line is calculated, which solves the accuracy and misidentification problems of identifying hidden danger targets in the prior art, and improves the safety of transmission line.

CN115995043BActive Publication Date: 2024-12-03SANLI VIDEO FREQUENCY SCI & TECH SHENZHEN
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310152840.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-02-16
Publication Date
2024-12-03
Estimated Expiration
2043-02-16

AI Technical Summary

Technical Problem

In the transmission line environment, it is difficult to accurately identify hidden danger targets below the skyline, and the error recognition rate is high, so high risk targets such as wildfires and smoke cannot be detected, and the edge detection effect is poor.

Method used

Deep learning-based object detection technology combined with edge detection, by detecting the transmission pole tower and the wire area, the physical distance between the candidate target and the wire is calculated, and whether it is a hidden danger target is determined based on the distance threshold.

Benefits of technology

It improves the accuracy of identifying hidden danger targets, reduces missed and false alarms, supports the detection of a variety of construction equipment and high-risk targets, and ensures the safety of transmission lines.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115995043B_ABST
    Figure CN115995043B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for identifying a target with hidden dangers in a power transmission line and a computer-readable storage medium. The method comprises: detecting a transmission tower and a detection frame of a candidate target in a power transmission image by a target detection technology based on deep learning; detecting a transmission wire in a power transmission image and determining a wire area; determining a distance threshold corresponding to the candidate target according to the type of the candidate target; calculating the physical distance between the candidate target and the wire area according to the actual height of the transmission tower, the height of the detection frame of the transmission tower, the position of the detection frame of the transmission tower and the detection frame of the candidate target in the power transmission image, and the imaging distance between the detection frame of the candidate target and the wire area; if the physical distance is less than or equal to the distance threshold corresponding to the candidate target, the candidate target is determined to be a hidden danger target. The present invention can improve the recognition accuracy of hidden danger targets.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of target detection technology, and in particular to a method for identifying targets of hidden dangers in power transmission lines and a computer-readable storage medium. Background Art

[0002] In power transmission scenarios, construction equipment or other objects often appear near power transmission lines, which poses certain risks to the safety of power transmission lines. Therefore, how to identify potential danger targets in power transmission scenarios is of great significance and can greatly improve the safety of power transmission lines.

[0003] A Chinese patent document with publication number CN107609556A proposes a method for detecting aerial work machinery in a power transmission line environment. The method first performs relevant preprocessing on the image, then uses edge detection to obtain its edge image, and then segments the sky area in the image based on the edge image. During detection, the edge image is differentiated to obtain a candidate detection area, and the background model is used to remove false hidden danger areas in the candidate area. Then, color information and edge curvature information are used to remove targets that do not meet the specified characteristics. Finally, the targets in the remaining candidate areas will be regarded as aerial work machinery and an alarm will be issued.

[0004] This method mainly uses edge detection to identify operating machinery above the skyline, but the edge detection effect is low. At the same time, this method cannot detect targets below the skyline that may threaten transmission lines. Furthermore, it does not support the detection of high-risk targets such as wildfires and smoke. In addition, it does not combine the transmission line to analyze whether the target will pose a threat to the safety of the transmission line, and there is a high misidentification rate. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a method for identifying hidden danger targets of a power transmission line and a computer-readable storage medium, which can improve the accuracy of identifying hidden danger targets.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: a method for identifying hidden danger targets of power transmission lines, comprising:

[0007] Through deep learning-based target detection technology, the detection frame of the transmission tower and the detection frame of the candidate target are detected in the transmission image;

[0008] Through edge detection technology, the power transmission wires are detected in the power transmission image and the wire area is determined;

[0009] Determine, according to the type of the candidate target, a distance threshold corresponding to the candidate target;

[0010] Calculate the physical distance between the candidate target and the wire area according to the actual height of the transmission tower, the height of the detection frame of the transmission tower, the positions of the detection frame of the transmission tower and the detection frame of the candidate target in the transmission image, and the imaging distance between the detection frame of the candidate target and the wire area;

[0011] If the physical distance is less than or equal to the distance threshold corresponding to the candidate target, the candidate target is determined to be a hidden danger target.

[0012] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method described above is implemented.

[0013] The beneficial effects of the present invention are as follows: by using a target detection algorithm based on deep learning, candidate targets in the transmission line scene can be identified with high integrity and high accuracy; by determining the conductor area and judging whether the candidate target is a hidden danger target based on the actual physical distance between the candidate target and the conductor area, it is possible to accurately distinguish whether the candidate target poses a threat to the transmission line, reduce the situation of missed reports and false alarms, and effectively improve the recognition accuracy of hidden danger targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A flow chart of a method for identifying hidden danger targets of power transmission lines according to the present invention;

[0015] Figure 2 This is a flow chart of a method according to Embodiment 1 of the present invention;

[0016] Figure 3 This is a schematic diagram of a power transmission image according to the first embodiment of the present invention;

[0017] Figure 4 The figure is a schematic diagram of the imaging principle of a camera according to the first embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is a detailed description in conjunction with the implementation methods and the accompanying drawings.

[0019] See also Figure 1 , a method for identifying hidden danger targets of power transmission lines, comprising:

[0020] Through deep learning-based target detection technology, the detection frame of the transmission tower and the detection frame of the candidate target are detected in the transmission image;

[0021] Through edge detection technology, the power transmission wires are detected in the power transmission image and the wire area is determined;

[0022] Determine, according to the type of the candidate target, a distance threshold corresponding to the candidate target;

[0023] Calculate the physical distance between the candidate target and the wire area according to the actual height of the transmission tower, the height of the detection frame of the transmission tower, the positions of the detection frame of the transmission tower and the detection frame of the candidate target in the transmission image, and the imaging distance between the detection frame of the candidate target and the wire area;

[0024] If the physical distance is less than or equal to the distance threshold corresponding to the candidate target, the candidate target is determined to be a hidden danger target.

[0025] From the above description, it can be seen that the beneficial effect of the present invention is that the recognition accuracy of hidden danger targets can be improved.

[0026] Furthermore, the edge detection technology is used to detect the power transmission wire in the power transmission image and determine the wire area, specifically:

[0027] By using a fast straight line detection algorithm, marginal wires in the power transmission wires are detected in the power transmission image, wherein the marginal wires include left marginal wires and right marginal wires;

[0028] Mapping the left marginal wire and the right marginal wire onto the horizontal ground respectively to obtain a left mapping line and a right mapping line;

[0029] The conductor area is determined according to the left mapping line, the right mapping line, the bottom line segment of the detection frame of the transmission tower, and the bottom line segment of the transmission image.

[0030] From the above description, it can be seen that by determining the conductor area, it is possible to effectively and accurately distinguish whether the candidate target poses a threat to the transmission line.

[0031] Further, the distance threshold corresponding to the candidate target is determined according to the type of the candidate target, specifically:

[0032] Determining the risk level of the candidate target according to the type of the candidate target;

[0033] According to the risk level of the candidate target, the distance threshold corresponding to the candidate target is determined.

[0034] From the above description, it can be seen that by grading the risk level of the target and setting the distance threshold according to the risk level, the recognition sensitivity of high-risk targets can be improved.

[0035] Further, the physical distance between the candidate target and the wire area is calculated according to the actual height of the transmission tower, the height of the detection frame of the transmission tower, the positions of the detection frame of the transmission tower and the detection frame of the candidate target in the transmission image, and the imaging distance between the detection frame of the candidate target and the wire area, specifically:

[0036] Determining the imaging height of the transmission tower according to the height of the detection frame of the transmission tower;

[0037] Acquire the actual height of the transmission tower, and calculate the scaling ratio corresponding to the bottom line segment of the detection frame of the transmission tower according to the imaging height and the actual height of the transmission tower as the first scaling ratio;

[0038] According to the height of the transmission image and the position of the bottom line segment of the detection frame of the transmission tower in the transmission image, calculating the imaging distance between the center height of the transmission image and the bottom line segment of the detection frame of the transmission tower as the first imaging distance;

[0039] Calculate, according to the height of the power transmission image and the position of the bottom line segment of the detection frame of the candidate target in the power transmission image, an imaging distance between the center height of the power transmission image and the bottom line segment of the detection frame of the candidate target as a second imaging distance;

[0040] Calculate, according to the first scaling ratio, the first imaging distance, and the second imaging distance, a scaling ratio corresponding to a bottom line segment of the detection frame of the candidate target as a second scaling ratio;

[0041] According to the position of the bottom line segment of the detection frame of the candidate target in the power transmission image and the position of the wire area in the power transmission image, calculating the imaging distance between the bottom line segment of the detection frame of the candidate target and the wire area in the horizontal direction as a third imaging distance;

[0042] The physical distance between the candidate target and the guide wire area is calculated according to the third imaging distance and the second scaling ratio.

[0043] From the above description, it can be seen that according to the camera imaging principle, the actual physical distance between the candidate target and the wire area is obtained by analysis and calculation.

[0044] Further, the actual height of the transmission tower is obtained, and according to the imaging height and the actual height of the transmission tower, the scaling ratio corresponding to the bottom line segment of the detection frame of the transmission tower is calculated as the first scaling ratio, which is specifically:

[0045] The actual height of the transmission tower is divided by the imaging height of the transmission tower to obtain a scaling ratio corresponding to a bottom line segment of the detection frame of the transmission tower, which is used as a first scaling ratio.

[0046] Furthermore, the scaling ratio corresponding to the bottom line segment of the detection frame of the candidate target is calculated as the second scaling ratio according to the first scaling ratio, the first imaging distance, and the second imaging distance, specifically:

[0047] The first scaling ratio is multiplied by the first imaging distance and then divided by the second imaging distance to obtain a scaling ratio corresponding to the bottom line segment of the detection frame of the candidate target as the second scaling ratio.

[0048] Further, the physical distance between the candidate target and the guide wire area is calculated according to the third imaging distance and the second scaling ratio, specifically:

[0049] The third imaging distance is multiplied by the second scaling ratio to obtain a physical distance between the candidate target and the guide wire area.

[0050] Furthermore, after determining that the candidate target is a potential risk target, the method further includes:

[0051] An alarm is issued according to the alarm strategy corresponding to the candidate target.

[0052] It can be seen from the above description that the safety of the transmission line is improved.

[0053] Furthermore, the candidate targets include but are not limited to construction machinery, wildfires and smoke, and the construction machinery includes but is not limited to tower cranes, forklifts, bulldozers, excavators, dump trucks, cranes, cement pump trucks, ship cranes, pile drivers and long-arm excavators.

[0054] From the above description, it can be seen that it can support the detection of various construction equipment, support the detection of wildfires and smoke, support the expansion of target types, and improve the comprehensiveness of target recognition.

[0055] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method described above is implemented.

[0056] Embodiment 1

[0057] Please refer to Figure 2-4 , Embodiment 1 of the present invention is: a method for identifying hidden danger targets of power transmission lines, which can be applied to power transmission scenarios. Figure 2 As shown, the following steps are included:

[0058] S1: Through deep learning-based target detection technology, the detection frame of the transmission tower in the transmission scene and the detection frame of the candidate target are detected in the transmission image.

[0059] like Figure 3 As shown, Figure 3 T0 is the detection frame of the transmission tower, and T1 is the detection frame of the candidate target.

[0060] In this embodiment, target detection technologies include but are not limited to target detection algorithms such as Yolo, RCNN, and SSD. Candidate targets include but are not limited to construction machinery, wildfires, and smoke. Construction machinery includes but is not limited to tower cranes, forklifts, bulldozers, excavators, dump trucks, cranes, cement pump trucks, ship cranes, pile drivers, and long-arm excavators.

[0061] S2: Using edge detection technology, the transmission line is detected in the transmission image and the line area is determined.

[0062] Specifically, the marginal wires in the transmission lines are detected in the transmission image by using the Fast Line Detector algorithm of Opencv, and the marginal wires include the left marginal wires and the right marginal wires; then the left marginal wires and the right marginal wires are respectively mapped to the horizontal ground to obtain the left mapping line and the right mapping line; finally, the wire area is determined according to the left mapping line and the right mapping line, the bottom line segment of the detection frame of the transmission tower and the bottom line segment of the transmission image.

[0063] like Figure 3 As shown, Figure 3 L1 in the figure is the left marginal wire, which is extended to intersect with the detection frame T0 of the transmission tower; L2 is the right marginal wire, which is extended to intersect with the detection frame T0 of the transmission tower; L1' is the mapping line segment of the left marginal wire on the horizontal ground, that is, the left mapping line, and the mapping basis is that the upper left starting point of L1 is mapped to the bottom of the transmission image, and the lower right end point of L1 is mapped to the bottom of the detection frame T0 of the transmission tower; L2' is the mapping line segment of the right marginal wire on the horizontal ground, that is, the right mapping line, and the mapping basis is that the upper right starting point of L2 is mapped to the bottom of the transmission image, and the lower left end point of L2 is mapped to the bottom of the detection frame T0 of the transmission tower; O is the intersection of the extension line of L1' and the extension line of L2'.

[0064] in addition, Figure 3 Where a is the lower left corner point of the detection frame T0 of the transmission tower, b is the lower right corner point of the detection frame T0 of the transmission tower, and line segment ab is the bottom line segment of the detection frame T0 of the transmission tower.

[0065] The conductor area is the area enclosed by L1', line segment ab, L2' and the bottom of the power transmission image.

[0066] S3: Determine a distance threshold corresponding to the candidate target according to the type of the candidate target.

[0067] In this embodiment, the risk level of the candidate target is first determined according to the type of the candidate target, and then the distance threshold corresponding to the candidate target is determined according to the risk level of the candidate target.

[0068] Among them, the risk level of candidate targets is graded according to the danger level and can be divided into low risk and high risk, as follows:

[0069] Low-risk targets: tower cranes, forklifts, bulldozers, excavators, dump trucks;

[0070] High-risk targets: cranes, cement pump trucks, ship cranes, pile drivers, long-arm excavators, forest fires, and smoke.

[0071] In this embodiment, the distance threshold for high-risk targets is 60 meters, and the distance threshold for low-risk targets is 30 meters.

[0072] S4: Calculate the physical distance between the candidate target and the wire area.

[0073] like Figure 4 As shown, assuming that the target is located at position P0, A0 is the vertex of the target, B0 is the bottom point of the target, O0 is the intersection of the camera imaging horizontal line and the target, and the points corresponding to the imaging of the target in the camera are A0', B0', and O0'; the actual height of the target is H0, the horizontal distance between the target and the camera is S0, the optical center of the camera is R, and the focal length is f. According to the camera imaging principle, the scaling ratio k0 of the A0' position in the image can be calculated, that is, k0=A0O0 / A0'O0', wherein A0O0 is the distance between point A0 and point O0 (physical distance), and A0'O0' is the distance between point A0' and point O0' (imaging distance).

[0074] Similarly, when the target is located at any position P, the scaling ratio of the position A' in the image is k=AO / A'O0'.

[0075] Since A0O0 is equal to AO, k0 / k=A'O0' / A0'O0'.

[0076] If the actual height of the target is known and the imaging height of the target can be obtained through the image, then the scaling ratio k0 of the P0 position can be calculated, and the ratio of A'O0' and A0'O0' can be calculated through the image, then the scaling ratio k of any position P can be calculated.

[0077] Combination Figure 3-4 , Figure 3Where c is the lower right corner of the candidate target's detection frame, e and f are the intersections of the extension of the bottom line segment of the candidate target's detection frame with the left mapping line L1' and the right mapping line L2', respectively, and d is the intersection of the line oc and the extension of the bottom line segment of the transmission tower's detection frame T0. Therefore, line segment ce represents the imaging distance between the bottom line segment of the candidate target's detection frame and the conductor area in the horizontal direction.

[0078] Assuming that the actual height of the tower is H0 and the imaging height is h0, the scaling ratio of the line segment ab at the bottom of the tower in the image is k0=H0 / h0.

[0079] Assuming that the height of the image (i.e., the power transmission image) is h, the ordinate of line segment ab in the image is y0, and the ordinate of line segment ce in the image is y (the origin of the image coordinate system is in the upper left corner of the image, and the positive direction of the Y axis is downward), we can get:

[0080] A0'O0'=|y0-0.5h|,

[0081] A'O0'=|y-0.5h|,

[0082] Therefore, the scaling ratio k at the line segment ce=k0*A0′O0′ / A′O0′=k0*(|y0−0.5h|) / (|y−0.5h|).

[0083] According to the imaging image, the length L of the line segment ce in the image can be calculated, so the actual distance of the line segment ce is S=L*k=L*k0*(|y0-0.5h|) / (|y-0.5h|).

[0084] Therefore, this step specifically includes the following steps:

[0085] S401: Determine the imaging height of the transmission tower according to the height of the detection frame of the transmission tower.

[0086] S402: Acquire the actual height of the transmission tower, and calculate the scaling ratio corresponding to the bottom line segment of the detection frame of the transmission tower according to the imaging height and the actual height of the transmission tower as the first scaling ratio.

[0087] For example, assuming that the imaging height of the transmission tower is 374 pix (pixels) and the actual height of the transmission tower is 50 m (meters), the first scaling ratio k0 = 50 / 374 = 0.134 m / pix.

[0088] S403: Calculate the imaging distance between the center height of the transmission image and the bottom line segment of the detection frame of the transmission tower according to the height of the transmission image and the position of the bottom line segment of the detection frame of the transmission tower in the transmission image as the first imaging distance.

[0089] S404: Calculate the imaging distance between the center height of the power transmission image and the bottom line segment of the candidate target's detection frame according to the height of the power transmission image and the position of the bottom line segment of the candidate target's detection frame in the power transmission image as the second imaging distance.

[0090] S405: Calculate the scaling ratio corresponding to the bottom line segment of the detection frame of the candidate target according to the first scaling ratio, the first imaging distance, and the second imaging distance as the second scaling ratio.

[0091] For example, assuming that the height of the transmission image is 1743pix, then the center height of the transmission image is 871pix. Assuming that the bottom line segment of the detection frame of the transmission tower ( Figure 3 The imaging distance between the line segment ab in the figure and the center height of the transmission image (i.e., the first imaging distance, i.e., Figure 4 A0'O0' in (A0'O0') is 341pix, and the imaging distance between the bottom line segment of the candidate target detection frame and the center height of the transmission image (i.e., the second imaging distance, i.e. Figure 4 A'O0' in is 491pix, then Figure 3 The scaling ratio at the middle line segment ce, that is, the second scaling ratio k=k0*A0'O0' / A'O0'=0.134*341 / 491=0.093 m / pix.

[0092] S406: Calculate the imaging distance between the bottom line segment of the candidate target detection frame and the wire area in the horizontal direction according to the position of the bottom line segment of the candidate target detection frame in the power transmission image and the position of the wire area in the power transmission image as a third imaging distance.

[0093] S407: Calculate the physical distance between the candidate target and the guide wire area according to the third imaging distance and the second scaling ratio.

[0094] For example, assuming that the third imaging distance (i.e., obtaining Figure 3 The length of the line segment ce in is 274pix, so the physical distance between the candidate target and the wire area is S=0.093*274=25.5m.

[0095] S5: Determine whether the physical distance is less than or equal to the distance threshold corresponding to the candidate target. If so, execute step S6.

[0096] S6: Determine that the candidate target is a potential risk target.

[0097] When a candidate target appears near the conductor area but has not yet touched the conductor area, it poses a threat to the transmission line. Therefore, when the physical distance between the candidate target and the conductor area is less than the corresponding distance threshold, the candidate target is considered to be a hidden danger target.

[0098] For example, Figure 3 The candidate target detected in the figure is smoke. Smoke is a high-risk target, and its corresponding distance threshold is 60 meters. The physical distance between the smoke detection frame and the wire area is 25.5 meters, which is less than the distance threshold of 60 meters. Therefore, it is considered that Figure 3 The smoke in the air is a hidden danger.

[0099] Furthermore, different warning strategies can be formulated according to the risk level of the hidden danger target and the physical distance from the conductor area.

[0100] For example, when the hidden danger target is a low-risk target, if the hidden danger target is more than 60 meters away from the wire area, a prompt alarm will be issued; if it is 40-60 meters away from the wire area, a third-level alarm will be issued; if it is 20-40 meters away from the wire area, a second-level alarm will be issued; if it is within 20 meters away from the wire area, a first-level alarm will be issued; if the hidden danger target is located in the wire area, an emergency alarm will be issued.

[0101] When the hidden danger target is a high-risk target, if the hidden danger target is more than 80 meters away from the wire area, a prompt alarm will be issued; if it is 60-80 meters away from the wire area, a third-level alarm will be issued; if it is 40-60 meters away from the wire area, a second-level alarm will be issued; if it is within 40 meters away from the wire area, a first-level alarm will be issued; if the hidden danger target is located in the wire area, an emergency alarm will be issued.

[0102] In this embodiment, by using a target detection algorithm based on deep learning, the accuracy of target detection can be effectively improved, and it can support the detection of various construction equipment, support the detection of wildfires and smoke, support the expansion of target types, and improve the comprehensiveness of target recognition; by determining the conductor area, it can effectively and accurately distinguish whether the candidate target poses a threat to the transmission line; by grading the risk level of the target and setting the distance threshold according to the risk level, the recognition sensitivity of high-risk targets can be improved; by judging whether the candidate target is a hidden danger target based on the actual physical distance between the candidate target detection box and the conductor area, it can effectively reduce missed reports and false alarms, and effectively improve the accuracy of hidden danger identification.

[0103] Embodiment 2

[0104] This embodiment is a computer-readable storage medium corresponding to the above embodiment, on which a computer program is stored. When the program is executed by the processor, it implements the various steps of a transmission line hidden danger target identification method in the above embodiment, and can achieve the same technical effect, which will not be repeated here.

[0105] In summary, the present invention provides a method for identifying hidden danger targets in transmission lines and a computer-readable storage medium. By using a target detection algorithm based on deep learning, the accuracy of target detection can be effectively improved, and the detection of various construction equipment, the detection of wildfires and smoke, and the expansion of target types can be supported to improve the comprehensiveness of target identification. By determining the conductor area, it is possible to effectively and accurately distinguish whether the candidate target poses a threat to the transmission line. By grading the risk level of the target and setting the distance threshold according to the risk level, the recognition sensitivity of high-risk targets can be improved. By judging whether the candidate target is a hidden danger target based on the actual physical distance between the candidate target detection box and the conductor area, it is possible to effectively reduce missed reports and false alarms, and effectively improve the accuracy of hidden danger identification.

[0106] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for identifying hidden danger targets of power transmission lines, characterized in that: include: Through deep learning-based target detection technology, the detection frame of the transmission tower and the detection frame of the candidate target are detected in the transmission image; Through edge detection technology, the power transmission wires are detected in the power transmission image and the wire area is determined; Determine, according to the type of the candidate target, a distance threshold corresponding to the candidate target; Calculate the physical distance between the candidate target and the wire area according to the actual height of the transmission tower, the height of the detection frame of the transmission tower, the positions of the detection frame of the transmission tower and the detection frame of the candidate target in the transmission image, and the imaging distance between the detection frame of the candidate target and the wire area; If the physical distance is less than or equal to the distance threshold corresponding to the candidate target, the candidate target is determined to be a potential hazard target; The edge detection technology is used to detect the transmission wires in the transmission image and determine the wire area, specifically: By using a fast straight line detection algorithm, marginal wires in the power transmission wires are detected in the power transmission image, wherein the marginal wires include left marginal wires and right marginal wires; Mapping the left marginal wire and the right marginal wire onto the horizontal ground respectively to obtain a left mapping line and a right mapping line; Determine a conductor area according to the left mapping line and the right mapping line, a bottom line segment of the detection frame of the transmission tower, and a bottom line segment of the transmission image; The calculating of the physical distance between the candidate target and the conductor area according to the actual height of the transmission tower, the height of the detection frame of the transmission tower, the positions of the detection frame of the transmission tower and the detection frame of the candidate target in the transmission image, and the imaging distance between the detection frame of the candidate target and the conductor area is specifically as follows: Determining the imaging height of the transmission tower according to the height of the detection frame of the transmission tower; Acquire the actual height of the transmission tower, and calculate the scaling ratio corresponding to the bottom line segment of the detection frame of the transmission tower according to the imaging height and the actual height of the transmission tower as the first scaling ratio; According to the height of the transmission image and the position of the bottom line segment of the detection frame of the transmission tower in the transmission image, calculating the imaging distance between the center height of the transmission image and the bottom line segment of the detection frame of the transmission tower as the first imaging distance; Calculate, according to the height of the power transmission image and the position of the bottom line segment of the detection frame of the candidate target in the power transmission image, an imaging distance between the center height of the power transmission image and the bottom line segment of the detection frame of the candidate target as a second imaging distance; Calculate, according to the first scaling ratio, the first imaging distance, and the second imaging distance, a scaling ratio corresponding to a bottom line segment of the detection frame of the candidate target as a second scaling ratio; According to the position of the bottom line segment of the detection frame of the candidate target in the power transmission image and the position of the wire area in the power transmission image, calculating the imaging distance between the bottom line segment of the detection frame of the candidate target and the wire area in the horizontal direction as a third imaging distance; The physical distance between the candidate target and the guide wire area is calculated according to the third imaging distance and the second scaling ratio.

2. The method for identifying transmission line hidden danger targets according to claim 1, characterized in that: The step of determining the distance threshold corresponding to the candidate target according to the type of the candidate target is specifically: Determining the risk level of the candidate target according to the type of the candidate target; According to the risk level of the candidate target, the distance threshold corresponding to the candidate target is determined.

3. The method for identifying transmission line hidden danger targets according to claim 1, characterized in that: The actual height of the transmission tower is obtained, and according to the imaging height and the actual height of the transmission tower, a scaling ratio corresponding to the bottom line segment of the detection frame of the transmission tower is calculated as the first scaling ratio, which is specifically: The actual height of the transmission tower is divided by the imaging height of the transmission tower to obtain a scaling ratio corresponding to a bottom line segment of the detection frame of the transmission tower, which is used as a first scaling ratio.

4. The method for identifying transmission line hidden danger targets according to claim 1, characterized in that: The step of calculating the scaling ratio corresponding to the bottom line segment of the detection frame of the candidate target as the second scaling ratio according to the first scaling ratio, the first imaging distance and the second imaging distance is specifically: The first scaling ratio is multiplied by the first imaging distance and then divided by the second imaging distance to obtain a scaling ratio corresponding to the bottom line segment of the detection frame of the candidate target as the second scaling ratio.

5. The method for identifying transmission line hidden danger targets according to claim 1, characterized in that: The calculating the physical distance between the candidate target and the guide wire area according to the third imaging distance and the second scaling ratio is specifically: The third imaging distance is multiplied by the second scaling ratio to obtain a physical distance between the candidate target and the guide wire area.

6. The method for identifying transmission line hidden danger targets according to claim 1, characterized in that: After determining that the candidate target is a potential danger target, the method further includes: An alarm is issued according to the alarm strategy corresponding to the candidate target.

7. The method for identifying hidden danger targets of power transmission lines according to any one of claims 1 to 6, characterized in that: The candidate targets include construction machinery, wildfires and smoke, and the construction machinery includes tower cranes, forklifts, bulldozers, excavators, dump trucks, cranes, cement pump trucks, ship cranes and pile drivers.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.