Target detection method for low-visibility power supply and distribution transformer area in foggy days

• Main Authors: BAI JINGTAO, XU TAO, ZHANG JIANPING, YAN ZHIPENG, LU KAI, LI CHONG, PENG KUN, LEI QIANG, ZOU HUIQUAN, LI HONGRI, WANG TENGDA, WANG ZEWEN, WU MINGCHAO, ZHENG MINGZHOU, WANG LEI, LONG PENG, ZHANG HAIRANG, TANG DONG, BAI JUN, JIE YEWEI, WU XINRUI, ZANG ZHI, CAO RUI, PENG BIN, ZHOU WENCHAO, FAN MING, CHEN DENGBO, SHEN JIANPING, LU WEIHONG, LIU SU
• Format: Patent
• Language: Chinese; English
• Published: 08.08.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention discloses a foggy-day low-visibility power supply and distribution area target detection method, and aims to solve the problem that target data extraction is affected due to the fact that a target image loses visual depth after defogging enhancement in a foggy-day low-visibility target detection method in the prior art. And obtaining a defogged and enhanced clear image, carrying out target detection on the clear image of the district by adopting a Yo-v4 algorithm, carrying out stereo matching by adopting an SGBM algorithm and a BM algorithm to obtain a visual depth image, and obtaining detection data of a standard district. The depth information of the standard court is extracted through a stereoscopic vision matching algorithm, the spatial three-dimensional coordinates of the detection points of the standard court are obtained, then the detection data of the standard court are calculated, the vision depth is recovered, and the data extraction precision is improved. Optimizing and updating the p