Method and system for detecting defects of power transmission line equipment

Abstract

The invention discloses a method and system for detecting defects of power transmission line equipment, and belongs to the technical field of power transmission line inspection. The method comprises:training by taking an equipment image as a training set to obtain a primary defect detection model; taking the generated defect recognition result data as a training set, and performing training to obtain a secondary defect detection model; and obtaining an inspection image of the target power transmission line as input data, inputting the inspection image into the primary defect detection model,obtaining an equipment image of the target power transmission line, taking the equipment image of the target power transmission line as secondary input data, inputting the secondary input data into the secondary defect detection model, and obtaining an equipment defect identification result of the target power transmission line. A two-stage detection and incremental detection model training mode is adopted, so that the defect recognition accuracy can be improved.

Description

technical field [0001] The present invention relates to the technical field of power transmission line inspection, and more specifically, to a method and system for detecting defects of power transmission line equipment. Background technique [0002] In recent years, unmanned aerial vehicles (UAVs) have been widely used in power inspections. They can collect images of transmission line equipment from multiple angles at short distances, which greatly improves the inspection efficiency compared with traditional manual inspections. At present, most inspection images in the industry rely on manual interpretation to find defects, which limits the further improvement of inspection efficiency. [0003] The background texture of the power equipment is complex, which makes it difficult to extract and distinguish the feature points of the power equipment, which affects the recognition effect. The normal equipment on the transmission line accounts for the vast majority, resulting in a...

AI Technical Summary

Problems solved by technology

At present, most inspection images in the industry rely on manual interpretation to find defects, which limits the further improvement of inspection efficiency

[0003] The background texture of the power equipment is complex, which makes it difficult to extract and distinguish the feature points of the power equipment, which affects the recognition effect

The normal equipment on the transmission line accounts for the vast majority, resulting in a high rate of false alarms in defect identification

At present, the recognition effect of mainstream deep learning network models cannot meet the requirements of industry applications

[0004] Traditional image recognition technology judges equipment defects through artificially designed matching templates, functions and other indicators, and cannot improve the recognition effect through incremental data

The recognition effect of the deep learning model can be improved with the expansion of the number of training sets. However, at present, manual labeling inspection image data is subject to the dual constraints of cost and efficiency, and the amount of data is difficult to achieve the theoretical detection accuracy of the existing deep learning network.

Images