Smart Solder Joint Visual Inspection

• Main Authors: Tsai-Pao Liu, 劉財寶
• Other Authors: Fang-Chin Tien
• Format: Others
• Language: zh-TW
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/233nr7

碩士 === 國立臺北科技大學 === 工業工程與管理系EMBA班 === 105 === In electronic manufacturing industry, the results of component soldering process directly affect product quality. Automatic Optical Inspection (AOI) evaluates the quality of manufactured products with the help of visual information. By using AOI system to control the quality of assembly, it is more effective and stable than traditional method that use operator’s experience to check out the defects. In this paper, focus on creating an AOI system that meet the requirement of factory. The experiment sample is fan’s PCB provided by D company. The printed circuit board (PCB) easily get dirty in process by soldering paste or other factors. It increased the risk of over kill by using traditional digital image process to detect defects. Artificial Neural Network (ANN) algorithm has good performance in solving complex problems. Trying to join the ANN algorithm in the system to reduce the risk of over kill and keep the inspection performance in the level.