Search

A support vector machine (SVM) is an important method in the detection and classification of the color difference on a polysilicon wafer. However, the accuracy of a SVM is affected by its feature vector and parameters. Owing to the complex color information and random texture features on the wafer surface, the feature design is extremely complicated. Meanwhile, a SVM optimized using a popular intelligent algorithm easily falls into a local optimum, and the convergence of the algorithm needs to be improved. Therefore, a classification method is proposed for detecting the color difference from multi-scale features in polysilicon wafer images. First, to extract the features, an image segmentation method is devised based on the maximum region contrast, which effectively applies a threshold segmentation of the wafer images. Second, the multi-scale features and color representations in different color spaces are used to construct a nine-dimensional feature vector that sufficiently describes the surface characteristics of the wafer. An approach to optimize the SVM is finally proposed using a magnetic bacteria optimization algorithm based on an elitist strategy for parameter optimization. The optimum individual of each generation is used to adjust the magnetic moment such that the solution approaches the optimal direction and enhances the global search ability. A fitness function is also introduced to improve the diversity of the solutions through a cross-validation method. The experiment results show that the proposed algorithm achieves an accuracy of 98.3% with a better classification performance than the other methods and that the color difference of polysilicon wafers can be effectively detected. [ABSTRACT FROM AUTHOR]

The recognition of color differences in solar cells with complex textures is a significant challenge in cell manufacturing. Traditional methods fail to detect the color difference effectively. Deep learning models have exhibited promise in many engineering fields. A multi-component attention-based convolution approach is proposed for the surface inspection based on the feature information in different color spaces. Wavelet entropy is employed to represent the information of different components, remove redundant components and extract effective feature information. Additionally, a residual attention mechanism is developed to capture local features with contextual semantic information. The best network structure is determined by evaluating the layer depth of the basic model and convolution kernel size. A multi-component network model is constructed based on the formed structure to improve the ability to distinguish different color difference features. Experimental results indicated that the proposed approach exhibits competitive performance. The research solution provides guidance for applications of deep learning to improve the quality of solar cells in manufacturing. [ABSTRACT FROM AUTHOR]