Development of a Machine Vision System for Solar Wafer Counting

The traditional manual counting wafers leaded to the silicon wafer cracked by operating frequently. Instead of the manual work, this paper proposed a system to counting wafers based on Machine Vision theory and Image Processing algorithm. We designed a counter system and adopted infrared led as parallel illumination source. In image pre-processing, this paper presented a series of algorithms, which contained image smoothing, uneven image correction and image morphology operation. This paper proposed a vertical projection counting based on statistics analysis substitute for the Hough straight lines detection, and the methods have achieved ideal effects by experimental results.