关键词: 表面检测, 缺陷检测, 缺陷分类, 质量控制

Abstract: Surface defect detection based on machine vision has been widely used in the various fields, which is important for ensuring the product quality during automatic production. However, there are some challenges on the surface inspection, such as low contrast, shape similarity between defect regions and non-defect regions, tiny defect detection, inspection speed and accuracy. So, this paper gives the recent advances in surface defect detection. The surface defect detection is categorized into three types: statistics, spectrum and model approach. Then this paper compares several typical approaches in detail, including feature extraction, detecting algorithms and the performance of the algorithms, and analyzes the effectiveness of the algorithms deeply. Finally, this paper summarizes the challenge and future trend.

Key words: surface detection, defect detection, defect classification, quality control