09-05-2023 дата публикации
Номер: CN116091395A
Автор:
WANG SIYUAN,
LI CONG,
LIU XIAO,
BAO XIN,
ZHANG YUWEI,
LI BIN,
ZHANG DECAI,
ZHOU XINGFU,
JIANG CHAO,
WEI LIANG,
LIU CHUNMING,
XIE XIANGYAN,
ZHENG YIBIN,
YANG JIE
Принадлежит:
The invention provides an electric power equipment oil leakage image identification method and system. The method comprises the steps of obtaining an original image of an electric power equipment oil leakage area; histogram equalization processing is carried out on the original image to enhance the contrast ratio of a detection area and a background in the original image, sharpening processing is carried out on the original image to enhance the edge part of the detection area, and a basic image is obtained after processing; and inputting the basic image into a semantic segmentation model to obtain information of the oil stain area, and generating a final image. Based on the electrical equipment oil leakage image recognition method, the invention further provides an electrical equipment oil leakage image recognition system. According to the method, the recognition result is further refined on the basis of the basic image through the semantic segmentation model, namely, pixel-level foreground ...
Подробнее