Surface oxidation defect detection of high-density flexible integrated circuit substrate

作者：Hu, Yueming; Li, Lu^*; Luo, Jiaxiang

来源：Computer Integrated Manufacturing Systems, 2020, 26(9): 2379-2387.

DOI：10.13196/j.cims.2020.09.007

• Adaptive thresholds
• Automatic Detection
• Color characteristics
• Color segmentation
• Manufacturing process
• Precision inspection
• Segmentation methods
• Surface oxidations

For the precision inspection requirement of high-density and ultra-thin flexible integrated circuit substrates, a detection method of oxidation defect was presented based on color characteristics. A fast segmentation method was designed for copper surface based on local histogram adaptive threshold. Then, a two-dimensional color segmentation method based on single Gauss model for detecting oxidation defects was proposed. An 8-dimensional color feature of oxidation defect was extracted based on RGB, HSV and block strategy, which input to the DAG-SVMS classifier to classify. The precision of method could reach micron level and the minimum detection area was 15 μm2. The proposed method provided a precise copper surface defect detection system for the manufacturing process of high-density and ultra-thin flexible integrated circuit substrate and solved the problem of automatic detection of oxidation defect on copper surface.

• 出版日期2020-9-1
• 单位华南理工大学