Application of Machine Vision to the Inspection of LED Lead Frame

• Main Authors: HAN,HUAN-JI, 韓煥奇
• Other Authors: chiou, yih-chih
• Format: Others
• Language: zh-TW
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/01715449850469054428

碩士 === 中華大學 === 機械工程學系碩士班 === 103 === It is important to discover any defect occurring on the surface of a LED lead frame. Undetected flaws might lower luminous efficiency or even fail to illuminate. In view of that we integrate stage control, lighting, and image processing techniques to develop an automated optical inspection system that is specifically designed to inspect LED lead frame surfaces for the following four kinds of flaws, namely, scratches, oil stains, foreign particles, and functional defected. To successfully detect the aforementioned flaws located at the functional region, we first search the image under test for the pre-defined pattern. After that we extract the pre-defined region of interested (ROI) from the test image. As soon as the desired ROIs were successfully segmented, a pre-defined flaw detection procedure is applied to check if there are any defects. Experimental results show that the proposed system can finish the inspection of a 1024 x 1024 image in 0.3 second and the flaw detection rate is 95.95%. In summary, the developed inspection system and method can not only replace the convention inspection method, but also increase the accuracy and efficiency of LED lead frame inspection.