An infrared image identification method and device for a lightning arrester

Abstract

The invention discloses an infrared image identification method of a lightning arrester. The method comprises the steps of obtaining an infrared image of a target lightning arrester; identifying and positioning a target lightning arrester in the infrared image through a preset deep learning model to obtain position information of the target lightning arrester in the infrared image, and training the deep learning model by adopting an SSD algorithm; and correspondingly storing the infrared image, the identification information of the target lightning arrester and the position information. According to the method, the position information of the lightning arrester in an infrared image is identified through a deep learning model; meanwhile, the infrared image, the identification information ofthe lightning arrester and the position information are correspondingly stored, and the corresponding marking process of the infrared image, the identification information of the lightning arrester and the position information is automatically realized, so that the marking efficiency and accuracy are improved, and reliable data is provided for the operation and maintenance inspection of the lightning arrester. Correspondingly, the invention also discloses an infrared image recognition device and equipment of the lightning arrester and a readable storage medium which also have the above technical effects.

Description

technical field [0001] The present invention relates to the technical field of infrared image processing, and more specifically, relates to an infrared image recognition method, device, equipment and readable storage medium of a lightning arrester. Background technique [0002] With the wide application of infrared equipment in live detection of arresters, power supply companies have accumulated a large number of infrared images of operation and maintenance of arresters. For the infrared images of these arresters, it is usually necessary to manually label the arresters in these infrared images. The information that needs to be marked generally includes: which arrester corresponds to which infrared image, where the arrester is located in the infrared image, and what tilt direction it has. The information obtained by marking can provide help for the operation and maintenance inspection of the arrester. Among them, the operation and maintenance inspection of the arrester gener...

AI Technical Summary

Problems solved by technology

[0003] However, due to the large number of lightning arresters, there are many corresponding infrared images, so the workload of labeling operations is relatively large.

Moreover, the efficiency of manual labeling is relatively slow, and errors and omissions will inevitably occur

Therefore, the existing technology has the disadvantages of low efficiency and difficult guarantee of accuracy

Images