Intelligent Instrument Recognition Scheme Based on Unattended Substation Inspection

Abstract

As the level of automation improves, unattended solutions become possible. Based on the substations with greater demand for intelligent inspection, this paper proposes a new instrument intelligent identification scheme, which has the following innovations: 1. Based on the structure of the inspection robot and the image acquisition process, this paper designs a digital substation pointer reading method based on digital image processing technology. 2. Analyze and study the methods of image graying, image smoothing, edge detection and line detection, especially the detailed research on Canny edge detection algorithm. 3. Combined with the unmanned substation patrol inspection practice, in view of the interference of complex lighting to the reading of the instrument, this paper proposes an adaptive Gamma enhancement algorithm based on Retinex theory. The details and colors of the image are binarized using a double threshold algorithm. The pointer instrument reading system designed in this paper has an average accuracy rate of 97%, which meets the actual requirements of inspection.