Color Recognition and Automatic Sorting System for Solar Cells of Microscope Imaging and Visual Detection based on HSI Color Space

  • Dong Lu
Keywords: Microscope Imaging; Visual Detection; Color Recognition; Color Classification Threshold; HSI Color Space


The color differences in solar cells are usually so small that they can only be detected using microscopic imaging
and visual detection techniques. In view of the color difference of solar cells, an automatic recognition and sorting
system based on microscope technique is designed in this paper. The method of high angle front lighting and low
angle side lighting is adopted to obtain the complete and clear image of the whole solar cells; the "V" type
conveyor belt in combination with the red hollow background plate is adopted to eliminate the color interference
of the background conveyor belt. The color classification threshold is calculated by taking samples from two
adjacent color transition regions; the main color region of the solar cells is extracted based on the gray histogram;
the color characteristic value of the solar cells is calculated based on the HSI color space through the tone value
H and the brightness value I, and the color type of the solar cells is determined based on the comparison between
the color characteristic value and the color classification threshold value. The experimental results show that the
detection error rate of the system is 0.17%, and the system can run for a long time without failure. Therefore, the
system can realize the color recognition and automatic sorting of the solar cells, and it is accurate and reliable,
which meets the requirements of production practice.