CN-122023428-A - Power transmission line sag recognition and early warning method and device combined with zooming ranging
Abstract
The invention discloses a power transmission line sag recognition early warning method and device combining zoom ranging, and relates to the technical field of power transmission line monitoring, wherein the method comprises the steps of installing a binocular camera system in a target recognition scene, adjusting parameters of a variable focus lens, calibrating the binocular camera system by utilizing a checkerboard, and constructing a parallax-depth mapping relation; the method comprises the steps of acquiring Q synchronous images by using a calibrated system, performing multi-point parallax analysis to obtain parallax analysis results of a first hanging point, a second hanging point and a sag minimum point, screening and determining analysis results of a target hanging point and the sag minimum point, taking the parallax analysis results as indexes, searching parallax-depth mapping relation to obtain depth information, calculating a target hanging point ranging result according to the depth information, calculating a sag value and judging whether the sag value exceeds a threshold value, and triggering early warning if the sag value exceeds the threshold value, so that the technical problem that potential safety hazards caused by sag overlarge cannot be effectively early warned due to the fact that sag change cannot be fed back in real time in a traditional ranging mode is solved.
Inventors
- LI XUEJUN
- WANG XIAOPENG
- DAI XIANGLONG
- JIANG YONG
- HE CHENGHU
Assignees
- 江苏濠汉信息技术有限公司
Dates
- Publication Date
- 20260512
- Application Date
- 20260415
Claims (8)
- 1. The power transmission line sag recognition and early warning method combined with zooming ranging is characterized by comprising the following steps of: installing a binocular camera system in a target recognition scene, performing parameter adjustment on a variable-focus lens in the binocular camera system, calibrating the binocular camera system subjected to focusing by using a checkerboard, and constructing a binocular parallax-depth mapping relation, wherein the binocular camera system is provided with the variable-focus lens; q times of synchronous image acquisition are carried out on the target identification scene by using the calibrated binocular camera system, Q synchronous image sets are obtained, and Q is an integer greater than or equal to 1; Traversing the Q synchronous image sets to perform multi-point parallax analysis to obtain Q first hanging point binocular parallax analysis results, Q second hanging point binocular parallax analysis results and Q sag lowest point binocular parallax analysis results, and respectively performing centralized screening on the Q first hanging point binocular parallax analysis results, the Q second hanging point binocular parallax analysis results and the Q sag lowest point binocular parallax analysis results to determine a target first hanging point binocular parallax analysis result, a target second hanging point binocular parallax analysis result and a target sag lowest point binocular parallax analysis result; Searching the binocular disparity-depth mapping relation by taking the target first hanging point binocular disparity analysis result, the target second hanging point binocular disparity analysis result and the target sag minimum point binocular disparity analysis result as indexes to obtain target first hanging point depth information, target second hanging point depth information and target sag minimum point depth information; Obtaining a target first hanging point distance measurement result, a target second hanging point distance measurement result and a target sag minimum point distance measurement result according to the target first hanging point depth information, the target second hanging point depth information and the target sag minimum point depth information; And calculating the sag value of the target identification scene based on the target first suspension point distance measurement result, the target second suspension point distance measurement result and the target sag minimum point distance measurement result, judging whether the sag value meets a preset sag threshold, and if not, obtaining first early warning information.
- 2. The method for identifying and early warning sag of a power transmission line in combination with zoom ranging according to claim 1, wherein traversing the Q synchronous image sets for multi-point parallax analysis to obtain Q first hanging point binocular parallax analysis results, Q second hanging point binocular parallax analysis results and Q sag lowest point binocular parallax analysis results comprises: Traversing the Q synchronous image sets to perform gray preprocessing to obtain Q preprocessed synchronous image sets; Extracting Q left preprocessed images in the Q preprocessed synchronous image sets to identify a first hanging point, a second hanging point and a sag lowest point, and obtaining Q left first hanging point positions, Q left second hanging point positions and Q left sag lowest point positions; Performing neighbor window identification in the Q left preprocessing images based on the Q left first hanging point positions, the Q left second hanging point positions and the Q left sag minimum point positions to obtain Q left first hanging point neighbor windows, Q left second hanging point neighbor windows and Q left sag minimum point neighbor windows; Based on the Q left first hanging point neighbor windows, the Q left second hanging point neighbor windows and the Q left sag minimum point neighbor windows, performing cost matching with Q right preprocessed images in the Q preprocessed synchronous image sets to obtain Q right matching first hanging point neighbor windows, Q right matching second hanging point neighbor windows and Q right matching sag minimum point neighbor windows; and carrying out one-to-one pixel value difference calculation on the Q left first hanging point neighbor windows, the Q left second hanging point neighbor windows and the Q left sag minimum point neighbor windows, and the Q right matching first hanging point neighbor windows, the Q right matching second hanging point neighbor windows and the Q right matching sag minimum point neighbor windows, and taking absolute values of calculation results as the Q first hanging point binocular parallax analysis results, the Q second hanging point binocular parallax analysis results and the Q sag minimum point binocular parallax analysis results.
- 3. The method for identifying and early warning sag of a power transmission line in combination with zoom ranging according to claim 2, wherein based on the Q left first hanging point neighbor windows, the Q left second hanging point neighbor windows, and the Q left sag minimum point neighbor windows, performing cost matching with the Q right preprocessed images in the Q preprocessed synchronization image sets to obtain Q right matching first hanging point neighbor windows, Q right matching second hanging point neighbor windows, and Q right matching sag minimum point neighbor windows, comprising: matching cost analysis is carried out on the Q left first suspension point neighbor windows and Q right first matching windows in the Q right preprocessed images, so that Q right first matching costs are obtained; Iterating the Q right first matching windows based on the sizes of the Q right first matching costs to obtain Q right second matching windows; performing matching cost analysis on the Q left first suspension point neighbor windows and the Q right second matching windows to obtain Q right second matching costs; Iterating the Q right second matching windows based on the sizes of the Q right second matching costs until the Q right preprocessed images are iterated, and respectively taking the matching windows corresponding to the minimum value of the matching costs in the iteration process as Q right matching first hanging point neighbor windows; and based on the Q left second hanging point neighbor windows and the Q left sagging lowest point neighbor windows, performing cost matching with the Q right preprocessed images to obtain Q right matching second hanging point neighbor windows and Q right matching sagging lowest point neighbor windows.
- 4. The method for identifying and early warning sag of a power transmission line in combination with zoom ranging according to claim 3, wherein iterating the Q right first matching windows based on the magnitudes of the Q right first matching costs to obtain Q right second matching windows comprises: Judging whether the Q right first matching costs are smaller than or equal to a preset cost value, if yes, moving the Q right first matching windows according to a first moving step length to obtain Q right second matching windows; If not, moving the Q right first matching windows according to a second moving step length to obtain Q right second matching windows, wherein the second moving step length is larger than the first moving step length.
- 5. The method for identifying and warning sag of a power transmission line by combining zoom ranging according to claim 1, wherein the steps of respectively performing centralized screening on the Q first hanging point binocular disparity analysis results, the Q second hanging point binocular disparity analysis results and the Q sag lowest point binocular disparity analysis results to determine a target first hanging point binocular disparity analysis result, a target second hanging point binocular disparity analysis result and a target sag lowest point binocular disparity analysis result include: calculating the average value of the Q first hanging point binocular parallax analysis results to obtain the average value of the first hanging point binocular parallax analysis results; taking the average value of the binocular parallax analysis results of the first hanging points as a starting point, and carrying out neighborhood construction in the Q first hanging point binocular parallax analysis results according to a preset neighborhood radius to obtain a starting neighborhood; Randomly extracting a first hanging point binocular parallax analysis result from the edge of the initial neighborhood as an iteration point, and carrying out neighborhood construction in the Q first hanging point binocular parallax analysis results based on a preset neighborhood radius to obtain an iteration neighborhood; Judging whether the neighborhood density of the iteration neighborhood is larger than or equal to the initial neighborhood, if so, randomly extracting a binocular parallax analysis result of a first hanging point from the edge of the iteration neighborhood to serve as an updating iteration point, continuously carrying out centralized screening based on the updating iteration point until the preset iteration times are met, and taking the updating iteration point obtained in the last iteration as a target first hanging point binocular parallax analysis result; And carrying out centralized screening on the Q second hanging point binocular disparity analysis results and the Q sagging lowest point binocular disparity analysis results, and determining a target second hanging point binocular disparity analysis result and a target sagging lowest point binocular disparity analysis result.
- 6. The method for identifying and pre-warning sag of a power transmission line in combination with zoom ranging according to claim 5, wherein determining whether the neighborhood density of the iterative neighborhood is greater than or equal to the initial neighborhood comprises: if not, a first random value and a second random value are obtained based on a random number generator, and when the first random value is larger than or equal to the second random value, a starting point is used as a target first hanging point binocular parallax analysis result; And when the first random value is smaller than the second random value, randomly extracting a first hanging point binocular parallax analysis result from the edge of the iteration neighborhood to serve as an updating iteration point, continuously carrying out centralized screening based on the updating iteration point until the preset iteration times are met, and taking the updating iteration point obtained in the last iteration as a target first hanging point binocular parallax analysis result.
- 7. The method for identifying and early warning the sag of the power transmission line by combining zooming ranging according to claim 1, which is characterized by further comprising the following steps: Continuously monitoring sag values in a preset monitoring window to obtain a continuously monitored sag value set; and when the fluctuation variance of the continuous monitoring sag value set is greater than or equal to a preset fluctuation variance, obtaining second early warning information.
- 8. The power transmission line sag identification and early warning device combined with zooming and ranging is characterized in that the device is used for implementing the power transmission line sag identification and early warning method combined with zooming and ranging according to any one of claims 1-7, and comprises the following steps: the binocular camera calibration module is used for installing a binocular camera system in a target recognition scene, carrying out parameter adjustment on a variable-focus lens in the binocular camera system, calibrating the binocular camera system subjected to focusing by using a checkerboard, and constructing a binocular parallax-depth mapping relation, wherein the binocular camera system is provided with the variable-focus lens; the synchronous image acquisition module is used for acquiring Q synchronous images of the target identification scene by using the calibrated binocular camera system to obtain Q synchronous image sets, wherein Q is an integer greater than or equal to 1; The parallax analysis screening module is used for traversing the Q synchronous image sets to perform multi-point parallax analysis, obtaining Q first hanging point binocular parallax analysis results, Q second hanging point binocular parallax analysis results and Q sagging lowest point binocular parallax analysis results, and respectively performing centralized screening on the Q first hanging point binocular parallax analysis results, the Q second hanging point binocular parallax analysis results and the Q sagging lowest point binocular parallax analysis results to determine a target first hanging point binocular parallax analysis result, a target second hanging point binocular parallax analysis result and a target sagging lowest point binocular parallax analysis result; The depth retrieval module is used for retrieving the binocular disparity-depth mapping relation by taking the target first hanging point binocular disparity analysis result, the target second hanging point binocular disparity analysis result and the target sag minimum point binocular disparity analysis result as indexes to obtain target first hanging point depth information, target second hanging point depth information and target sag minimum point depth information; The distance calculation module is used for obtaining a target first hanging point distance measurement result, a target second hanging point distance measurement result and a target sag minimum point distance measurement result according to the target first hanging point depth information, the target second hanging point depth information and the target sag minimum point depth information; And the early warning module is used for calculating the sag value of the target identification scene based on the target first suspension point distance measurement result, the target second suspension point distance measurement result and the target sag minimum point distance measurement result, judging whether the sag value meets a preset sag threshold, and if not, acquiring first early warning information.
Description
Power transmission line sag recognition and early warning method and device combined with zooming ranging Technical Field The application relates to the technical field of power transmission line monitoring, in particular to a power transmission line sag recognition early warning method and device combining zooming ranging. Background During operation of a power transmission line, sag problems have been an important factor affecting power transmission efficiency and safety. Sag refers to the phenomenon that a power line wire sags under the action of dead weight, and particularly in a long-distance and high-voltage power transmission line, the sag is too large, so that the safety distance between the wire and the ground or other objects is reduced, and the risks of safety accidents such as short circuit, electric shock and the like are increased. In addition, the excessive sag can also cause mechanical damage to the transmission line, thereby affecting the stability and efficiency of power transmission. The traditional sag detection method is mostly dependent on fixed position monitoring equipment, and has the limitations of insufficient precision, poor real-time performance and difficult coverage of the whole power transmission line. Under complex environmental conditions, such as weather changes, complex circuit layout and the like, the sag monitoring means in the prior art cannot meet the real-time and accurate requirements, cannot timely send out risk early warning, and brings potential safety hazards to a power system. Disclosure of Invention By providing the power transmission line sag identification early warning method and device combined with zooming ranging, the technical problem that sag change cannot be fed back in real time in a traditional ranging mode, so that potential safety hazards caused by excessive sag cannot be effectively early warned is solved, and the technical effects of accurately monitoring sag change in real time in a mode of combining a binocular camera system with zooming ranging and improving safety and efficiency of a power transmission line are achieved. The application provides a power transmission line sag identification early warning method combining zoom ranging, which comprises the steps of installing a binocular camera system under a target identification scene, carrying out parameter adjustment on a variable-focus lens in the binocular camera system, calibrating the focused binocular camera system by using a checkerboard, constructing a binocular parallax-depth mapping relation, carrying out Q-time synchronous image acquisition on the target identification scene by using the calibrated binocular camera system, obtaining a Q-time synchronous image set, wherein Q is an integer greater than or equal to 1, traversing the Q-time synchronous image set, carrying out multi-point parallax analysis, obtaining a Q-first hanging point binocular parallax analysis result, a Q-second hanging point binocular parallax analysis result and a Q-time binocular parallax analysis result, respectively carrying out centralized screening on the Q-first hanging point binocular parallax analysis result, the Q-second hanging point binocular parallax analysis result and the Q-time binocular parallax analysis result, determining that the first hanging point binocular parallax analysis result is the first hanging point binocular parallax analysis result, the second hanging point binocular parallax analysis result is the second hanging point binocular parallax analysis result and the Q-depth of the second hanging point, determining that the first hanging point binocular parallax analysis result is the second hanging point binocular parallax analysis result is the target depth, and the target depth analysis result is the second hanging point binocular parallax analysis result, and the target depth is the target depth, and the second hanging point binocular parallax analysis result is the target depth, and the target depth is the second hanging point depth and the target depth. The method comprises the steps of obtaining a target first suspension point distance measurement result, a target second suspension point distance measurement result and a target sag minimum point distance measurement result, calculating a sag value of a target identification scene based on the target first suspension point distance measurement result, the target second suspension point distance measurement result and the target sag minimum point distance measurement result, judging whether the sag value meets a preset sag threshold, and if not, obtaining first early warning information. The application also provides a power transmission line sag recognition early warning device combining zooming ranging, which comprises a binocular camera calibration module, a binocular parallax-depth mapping relation and a power transmission line sag recognition early warning module, wherein the binocular camera calibration module is used for