Light source optimization scheme for paper defect detection system based on the uniform illumination of the near-field

The image collected in the paper defect detection system has lateral gray unevenness. Based on the theory of geometrical optical irradiance, this paper studies the application technology around the uniform distribution of illuminance on the paper detected by the light source module of the paper defect detection system. According to the Sparrow's criterion, the optimal distance function between the LED beads in the light source module is determined. From the perspective of machine vision, the power optimization function of the lamp bead at different positions in the light source array is determined by experimental method. The improvement effect of the light source was verified by using the image gray uniformity index. The experimental results show that the optimization scheme can increase the gray uniformity of the acquired image to 98.61%, which is greatly improved compared with 57.80% before optimization. This effectively improves the image quality of the paper and reduces the difficulty of image processing in the later stage, which is expected to improve the execution efficiency of the application.