Fast Automatic Registration Control and Inspection for Pre-Bond/ Main-Bond Machine in TFT LCD Manufacturing Process

• Main Authors: Wu-Kuo Chang, 吳國彰
• Other Authors: Lin. Chih-Min
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/47503151113181431155

碩士 === 逢甲大學 === 資訊電機工程碩士在職專班 === 96 === This research examines the electrical conductive particles of anisotropic conductive film (ACF) during the laminate manufacturing process for Chip On Glass(COG) of liquid crystal displays (LCD) and adopting an improved pattern match method to applying on-line automatic and relevant measurement inspections. This research uses pattern setup and analysis to add a matrix of miscellaneous gray scale to an adaptive one and run a miscellaneous algorithm on part of the pattern block to improve executive efficacy of the system. In addition, analysis and setup of the adaptive feature weight matrix is also introduced to reduce error match. In the pattern search, gradual proximity is used to adjust and determine the suitability of the searched objective pattern and multiple matches. Identification is conducted on a test image by plural block patterns and conductive particles patterns. The focus of this research is describing and identifying feature images with pattern matching technology. A novel mthod of genetic and back-propagation network(BPN) algorithm is presented in our system. We aim to improve traditional pattern match methods including the gray scale designs, adaptive pattern matrix, and adaptive feature weight pattern to reduce system function errors and to enable more efficient and quicker pattern searches and matches for full-size images.