Research on Classification Method of Vehicle Body Surface Defects Based on BP Neural Network. Issue 1 (January 2020)

• Record Type: Journal Article
• Authors: Wang, Xing; Wu, Mingming; Zhou, Jinxia; Wang, Jianguo
• Is Part Of: IOP conference series. Volume 740:Issue 1(2020)
• Publication Date: 2020-01
• Subjects: Materials science -- Periodicals; 620.1105
• DOI: 10.1088/1757-899X/740/1/012036
• Languages: English
• ISSNs: 1757-8981

Abstract

Abstract: In order to classify the surface defects of the vehicle body, a classification method based on BP (back propagation) neural network is proposed. According to the characteristics of vehicle body defects, the gray, morphological and geometric features of various defects were extracted, and the defect feature database was established. The defect classifier was designed by using improved BP neural network algorithm. The experimental results show that the method is very feasible and effective in the identification of vehicle body defects.