Improved multi-instrument reading identification method of transformer station inspection robot

Abstract

The invention discloses an improved multi-instrument reading identification method of a transformer station inspection robot. In the method, first of all, for instrument equipment images of different types, equipment template processing is carried out, and position information of min scales and max scales of each instrument in a template database. For the instrument equipment images acquired in real time by the robot, a template graph of a corresponding piece of equipment is scheduled from a background service, by use of a scale invariant feature transform (SIFT) algorithm, an instrument dial plate area sub-image is extracted in an input image in a matching mode, afterwards, binary and instrument point backbone processing is performed on the dial plate sub-image, by use of rapid Hough transform, pointer lines are detected, noise interference is eliminated, accurate position and directional angel of a pointer are accurately positioned, and pointer reading is finished. Such an algorithm is subjected to an on-site test of some domestic 500 kv intelligent transformer station inspection robot, the integration recognition rate of various instruments exceeds 99%, the precision and robustness for instrument reading are high, and the requirement for on-site application of a transformer station is completely satisfied.

Description

technical field [0001] The invention relates to an improved multi-instrument reading recognition method of a substation inspection robot. Background technique [0002] Substation equipment inspection robot is an outdoor all-weather mobile platform based on autonomous navigation, precise positioning, and automatic charging, integrating visible light, infrared, sound and other sensors; based on magnetic trajectory and specially arranged RFID tags on the road surface, the optimal path planning of the inspection robot is realized and two-way walking, and transmit the video, sound and infrared temperature measurement data of the detected equipment to the monitoring room through the wireless network; the inspection background system uses equipment image processing and pattern recognition technologies, combined with equipment image infrared expert database, to realize thermal monitoring of equipment. Discrimination of defects, opening and closing status, and abnormal appearance, as...

AI Technical Summary

Problems solved by technology

2003, 617–622] mentioned a comparative study of edge detection, and proposed an edge extraction method combining gradient operators and morphological operations, using fast Hough transform to extract circular or elliptical instrument panels, although its algorithm has certain Real-time and robust, but not suitable for substation instrument images collected by inspection robots

2003, 24 (4): 430–431], it is mentioned that the scale is regarded as a small straight line, and the pointer and the scale are recognized simultaneously by using the Hough transform. Although certain accuracy and calculation time can be guaranteed, for some multi-symbol, multi-stripe, etc. Meters that interfere do not apply

However, this algorithm has many preconditions and constraints. First, the zero-scale meter pointer image using the subtraction method must be on the same shooting plane as the input image. If there is an angle difference between the two images, it will affect the final pointer reading; secondly, the dial The scale image needs to be taken at close range and has a certain clarity, otherwise it will affect the recognition result of the scale, so it is not suitable for the recognition of instrument images taken by inspection robots in long-distance, multi-angle and multiple environments

Images