Application of computer vision technology to identification of feeder trays in integrated circuits

• Main Authors: I-CHUN TAI, 戴翊鈞
• Other Authors: 陳智勇
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/9hfq2g

碩士 === 樹德科技大學 === 電腦與通訊系碩士班 === 106 === The main purpose of this paper is to apply computer vision to identify whether there is a shortage of IC load in the tray. This method can assist the semiconductor industry, replacing manual inspections to reduce the odds of abnormalities and improve product yield. In this study, the captured images were first captured by feature points, and a complete tray image was formed. The histogram equalization was used to improve image quality. The image after equalization of the histogram is subjected to noise removal and edge detection processing. Finally, the presence or absence of IC in the material tray is identified through the image matching method. This study used 5 sets of ICs of different specifications and each group took 3 types of images for a total of 15 images to evaluate the effiency of the method. From the experimental results, it can be seen that computer vision application identification can achieve considerable accuracy in all kinds of ICs, and can also reduce the time for using manual inspections. It is quite suitable for applications in the semiconductor industry to reduce labor costs and time but also . Improve product yield rate.