Target detection and defect identification method for stockbridge damper of power transmission line

Abstract

The invention discloses a target detection and defect identification method for a stockbridge damper of a power transmission line. The method comprises the following steps: acquiring a stockbridge damper picture through an image acquisition device; preprocessing the shockproof hammer picture to obtain a preprocessed picture; constructing a feature extraction module, a feature enhancement module and a pixel-level prediction module; performing connection optimization on the feature extraction module, the feature enhancement module and the pixel-level prediction module, and establishing a training model; and inputting the pre-processed picture into a training model for testing, thereby realizing target detection and identification of the shockproof hammer in the pre-processed picture. The invention also provides a target detection and defect identification device for the shockproof hammer of the power transmission line. According to the invention, the detection speed of a high-resolutionpicture is greatly improved; calculation is reduced to a certain extent, rapid and accurate shockproof hammer target detection and defect identification are realized, and the efficiency of electric power inspection work is improved.

Description

technical field [0001] The invention relates to the technical field of safety of overhead power transmission lines, in particular to a method and device for target detection and defect identification of anti-vibration hammers of power transmission lines. Background technique [0002] In the overhead transmission line, the transmission line vibrates with the wind, and long-term periodic vibration can easily lead to strand breakage, which seriously threatens the safe and reliable operation of the overhead transmission line. The anti-vibration hammer is an important device in the overhead transmission line, which can reduce the vibration damage of the overhead transmission line. However, the overhead transmission line implements long-distance power transmission, and it will inevitably span different natural environments during the period. After being exposed to the sun for a long time, the anti-vibration hammer will have various defects, resulting in the anti-vibration hamme...

AI Technical Summary

Problems solved by technology

[0007] 2. Predefined anchors need to be redesigned on new detector tasks with different object sizes or aspect ratios

[0009] 4. Excessive anchors can also significantly increase computation and memory usage when computing the intersection-over-union ratio (IOU) between all anchors and ground-truth boxes during training

[0010] However, the network structure is relatively complex and there are relatively many parameters, so the calculation speed will be slightly slower and the memory space will be more occupied, which cannot meet some occasions with limited computing resources or high real-time requirements.

There is still a certain gap in the detection accuracy and efficiency of the anti-vibration hammer and its defects. When it is put into practical application, the existing problems will become more obvious.

Images