Cable part automatic identification method based on infrared image

• Main Authors: HU XINXIN, WEI YILONG, PEI XINGYU, LI JIAN, LIAO YANQUN, SHI XIAOCHUAN, ZOU YUBIN, HUANG JIAMING, ZHOU HAO, YE YUTING
• Format: Patent
• Language: Chinese; English
• Published: 21.01.2020
• Subjects: CALCULATING; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention relates to the technical field of power equipment maintenance, in particular to a cable part automatic identification method based on an infrared image, which comprises the steps of S1,constructing a training sample library of different parts of a cable; s2, reading training sample infrared images of different parts of the cable; s3, extracting training sample characteristic quantities of different parts of the cable and corresponding part labels; s4, training a support vector machine classifier by using the training sample; s5, reading infrared images of different parts of theto-be-tested cable; s6, extracting characteristic quantities of the infrared images of different parts of the to-be-tested cable; and S7, utilizing the trained support vector machine classifier and the extracted to-be-tested infrared image characteristic quantity to enable the to-be-tested infrared image to correspond to different part labels. According to the invention, the dependence on manual analysis and processing is