Defect detection method for solar cells based on human visual characteristics

Aiming at the problem that the defects of solar cells are diverse and difficult to detect, a detection method for surface defects of solar cells based on human visual characteristics was presented. Inspired by human visual characteristics, firstly, the line segment detector (LSD) was used to remove the grids that influence the defect detection, and then the Gabor filter texture suppression algorithm was proposed for texture suppression. Finally, a threshold segmentation based on the Chebyshev's theorem was proposed, and the control limit was set by the principle of statistical process control to divide the image pixels to realize the detection of surface defects of solar cells. Experimental results show that the proposed method is feasible in solar cells defect detection, it is effective and has a high detection rate.