Image Defect Detection of Solar Cell Based on Optimized GoogLeNet

Bowen Shao

Abstract: In order to deal with the energy crisis, people focus on renewable energy. Solar energy is playing an increasingly important role. As the core component of photovoltaic power generation, the flaws of solar cells will lead to low utilization of solar energy. Therefore, it is very important to detect the defects of solar cells. In this paper, an optimized GoogLeNet model is used to detect the image defects of solar cells. By adjusting the network structure and optimizing the activation function, the accuracy of the model reached 95.92%.

Keywords: solar cell, deep learning, defect detection, convolutional neural network