Surface Defect Recognition of Strip Steel Based on Fuzzy Broad Learning System
Konferenz: MEMAT 2022 - 2nd International Conference on Mechanical Engineering, Intelligent Manufacturing and Automation Technology
07.01.2022 - 09.01.2022 in Guilin, China
Tagungsband: MEMAT 2022
Seiten: 4Sprache: EnglischTyp: PDF
Autoren:
Teng, Zhi; Liang, Riqiang; Cheng, Hao; Xu, Zengmin (School of Mechanical and Electrical Engineering, Guilin University of Electronic Technology, China)
Wang, Guangbin; Jiang, Zhansi (School of Mechanical and Electrical Engineering, Guilin University of Electronic Technology, China & School of Mechanical and Electrical Engineering, Lingnan Normal University, China)
Inhalt:
The Fuzzy Broad Learning System(F-BLS) has a short calculation time, and on this basis can effectively alleviate the rule explosion problem in the neuro-fuzzy model. It is further improved from the Broad Learning System (BLS) and maintains the BLS network structure. This paper proposes a method that combines feature extraction and feature selection with fuzzy width learning. By using this structure to identify the surface defects of strip steel, the results prove that the final accuracy rate can reach 98.26% and the training time for each picture only needs 0.023521s. This not only proves the feasibility of the method but also further illustrates the flexibility of the fuzzy width learning system.