Solar cell defect detection method integrating short-time and long-time depth characteristics

Abstract

The invention provides a solar cell defect detection method integrating short-time and long-time depth characteristics. The method comprises the following steps: preprocessing; performing short-time depth feature extraction, including blocking and vectorizing the preprocessed images and then sending the images into a stacked noise reduction automatic encoder to be trained; obtaining a two-dimensional adaptive depth feature matrix learned by all image blocks, and converting the two-dimensional adaptive depth feature matrix into a three-dimensional matrix to obtain a short-time depth feature composed of current image observation information; extracting long-time depth features; integrating and converting the short-time depth feature and the long-time depth feature; and performing low-rank matrix decomposition and post-processing to obtain a final detection result. According to the method, the defect of the solar cell is characterized by using the depth characteristics fusing the currentimage observation information and the priori knowledge, so that the universality and accuracy of the defect detection of the solar cell can be remarkably improved, the calculation amount is small, thedetection efficiency is high, and the positioning precision is relatively high.

Description

technical field [0001] The invention relates to the technical field of solar cell defect detection, in particular to a solar cell defect detection method that combines short-term and long-term depth features. Background technique [0002] In recent years, facing the problems of environmental degradation and energy shortage, the use of renewable energy such as solar energy to generate electricity has become a key technical means to solve this problem, and as the key equipment of solar photovoltaic power generation systems, the quality of solar cells has a great has far-reaching consequences. Therefore, it is of great significance to detect the defects of the produced solar cells. [0003] There are many existing detection methods for solar cell defects, among which, the detection method based on machine vision has become the mainstream of current research because of its high efficiency and convenience. As an important link in the defect detection process, feature extraction...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the technical problems of poor versatility and insufficient detection accuracy of existing detection methods that simply use current image observation information or purely use prior knowledge, this invention proposes a solar cell defect detection method that combines short-term and long-term depth features , which effectively uses the deep features that fuse the current image observation information and prior knowledge to characterize the defects of solar cells, which can significantly improve the versatility and accuracy of single crystal silicon solar cell defect detection, and has a small amount of calculation and high detection efficiency

Images