Traceability Management Studies Crystal Silicon Bar Code Recognition Technology

Photovoltaic power generation is the most promising new energy technology developed recently,the use of solar panels generate electricity can provide a steady stream of clean energy for mankind. Crystal silicon as the basic main items for solar panels, battery performance is critical. Crystal silicon whether internal or external defects contain impurities, that affect photoelectric conversion efficiency and stability of the battery, so the need for testing before finishing products and traceability to ensure product quality. At present, domestic and foreign manufacturers of solar panels in the silicon crystal traceability is still in the research stage, the practical application in the blank. To solve this problem, first identified the bar code on the surface of the laser ingot, then cut along the direction perpendicular to the bar code, the wafer is cut into the side of the crystal are left bar code image, and finally through acquisition, image processing, image recognition bar code, from a growth of crystalline silicon ingots to the screening, packaging and various process aspects of traceability management.this paper designed a tiny bar code image capture device, through the CCD camera installed in a closed chassis, under the irradiation of near-infrared light to 170μm thickness of crystalline silicon side static bar code image capture. Then design software system, correction and image processing recognition bar code image position to achieve a crystalline silicon process traceability information.In the bar code image processing,at first,by analyzing and comparing several classical edge detection algorithms, sobel operator edge draw the best effect. Then according to the original bar code image capture obtained numerical size and stripe width strip empty edge cases, drew the template of improved sobel operator.At last, combined with the filtering, projection, filling processing technology to achieve thedetection of the bar code image.At same time,analyze the effect of the global threshold for bar code image processing in the bar code image processing and image post-processing sub-templates and improved sobel operator weighted integration. Considering the threshold and the local threshold to the acquisition of the bar code image edge and the whole process in different way, in image processing of detection regionis judged according to the edge of space position,different pixels, different ways of selecting two threshold values.Put forward combining the improved sobel operator image with two value method.Based on the above two types of bar code image processing methods, the refinement and smoothing the bar code image edge.Through the preparation of the bar code image restoration processing and recognition program, the establishment of the bar code information database, and developed a traceability management system platform, the research methods of screening for each step of the process traceability management form crystalline silicon ingots to wafers.