This paper analyzes the limitations of the traditional manual inspection of BIW surface defects,reveals the necessity of developing a visual inspection system for surface defects,introduces the composition and technical requirements of the detection system,constructs a method of surface defect data collection,creates a deep learning method based on multi-source data fusion to realize the detection and classification of BIW surface defects,and uses the built PIW visual inspection experimental platform for algorithm model training and method verification,and the overall consistency rate between manual inspection and machine inspection reaches 97.1%to achieve accurate detection and grading.
关键词
白车身/多源数据融合/缺陷检测/分级
Key words
Body-in-white/Multi-source Data Fusion/Defect Detection/Grading
维普
万方数据