Pyramid and loss function enhanced power system insulator and defect detection network

• Main Authors: XU ZHENGHUA, WANG BO
• Format: Patent
• Language: Chinese; English
• Published: 07.04.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention discloses a pyramid and loss function enhanced power system insulator and defect detection network, and particularly relates to the technical field of power systems, and the specific method comprises the following steps: S1, establishing an insulator defect depth detection system; s2, dividing the data set; s3, fusing multi-layer feature information by adopting a feature pyramid structure; s4, adopting a CIoU Loss calculation mode for a loss function; s5, calculating a classification network branch loss constraint training process; s6, calculating a regression loss network branch loss constraint training process; and S7, inputting the divided data set in the step S2 into the pyramid and loss function enhancement-based insulator and defect depth detection network obtained in the step S6, and automatically obtaining a detection effect picture. The insulator and defect depth detection network based on feature pyramid and loss function enhancement provided by the invention has good detection precisi