A New Auto-Inspection System for SMD PCB by Vision Inspection Technique

• Main Authors: Guang-Yu Peng, 彭光裕
• Other Authors: Der-Baau Perng
• Format: Others
• Language: zh-TW
• Published: 2000
• Online Access: http://ndltd.ncl.edu.tw/handle/61663497222649217806

碩士 === 國立交通大學 === 工業工程與管理系 === 88 === The SMT Machine in PCB Assembly is moving up rapidly during the past decades. While many defects, such as component missing, misalignment, wrong polarity, and solder bridge, still hide in the working process. This paper devised and developed a new system by utilizing the image processing technique including Correlation Coefficient Algorithms, Histogram, Morphology, and Blob process to inspect the above defects automatically. The SMD components of resistors, chip capacitors, SOP and QFP are mainly focused. A new lighting environment comprises three-layer of ring-like LED Array is also designed to provide appropriate light for grabbing the desired PCB image. Proper lighting environment for inspecting the defects of different components is discussed in this paper. A three-tier inspection architecture is devised in this new system, which includes: (1) Standard component library construction module; (2) On-Board component setup module; and (3) In-Line inspection module. By using this architecture, we can simplify the procedure of training and inspection. Experimental results show that the proposed method is robust for SMT PCB inspection.