Solar cell sorting method and system based on machine vision

Abstract

The invention relates to a solar cell color difference sorting method and system based on machine vision. The solar cell color difference sorting method comprises the steps of collecting images of to-be-sorted cells; dividing the images into a plurality of subimages which are crossed vertically and horizontally, calculating an LAB three-component average value of each subimage and color difference values between the subimages, and finding out the maximal color difference value of the subimages; judging whether the maximal color difference value is greater than a set threshold, if the maximal color difference value is greater than a set threshold, cell colors are not uniform, a detection result is directly output and is sent to a sorting control module for the next operation, and image processing is finished; if the maximal color difference value is not greater than the set threshold, the cell colors are uniform, and the color difference between a to-be-tested image and each standard image is calculated according to the LAB three-component average values of the to-be-sorted images and different colors of standard images so as to judge the color type of the cells. According to the solar cell color difference sorting method, the automatic solar cell color difference sorting is realized, and the sorting quality and efficiency are increased.

Description

technical field [0001] The invention relates to the field of solar photovoltaics, in particular to a machine vision-based color difference sorting system for solar cells. Background technique [0002] There will be a coating process in the production process of solar cells, and the difference in the coating process will lead to differences in the color of the cells. In the existing solar photovoltaic industry, there are not only color differences inside the cells, but also color differences between the cells. The chromatic aberration problem of the cells is particularly serious, leading to an increasing number of customer complaints. In order to solve this problem, the solar cells are generally inspected and analyzed manually, and the cells with color difference in the sorting place; but this process consumes too much human resources, the accuracy of manual inspection is not high, and manual long-term inspection will cause fatigue and affect the sorting effect. , Poor sorti...

AI Technical Summary

Problems solved by technology

For example, a color difference automatic sorting device for solar cells disclosed in the application document CN203292095U detects whether there is a color difference between the cells according to multiple sensors, and sorts the ones that meet the requirements; The method for automatic detection and sorting of cells and the detection and sorting of cells with uniform single color in the prior art; the technology in the prior art is the premise of color difference sorting of cells All cells have uniform monochromatic color, and it is impossible to detect solar cells with uneven color inside the cells, and it is impossible to sort out cells with uneven colors

Images