基于支持向量机分类策略的多晶硅电池片色差检测

针对复杂颜色和纹理特征条件下,多晶硅电池片上的色差检测问题,提出了一种基于支持向量机分类策略的多晶硅电池片色差检测方法。首先对预处理后电池片图像进行颜色模型转换和通道分离,利用Otsu方法对单通道图像进行阈值分割处理,并计算各阈值图像的区域对比度,然后根据区域对比度情况选择合适的阈值图像,利用阈值图像所提供的信息提取图像特征;最后使用支持向量机分类器来判别电池片是否存在色差缺陷。实验结果表明提出的色差检测算法可以实现多晶硅电池片色差高效检测,色差缺陷检测的准确度、误检率和检测时间分别达到96.88%, 5％和109ms。

Color Difference Detection of Polycrystalline Silicon Cells Based on Support Vector Machine Classification Strategy

Aiming at the problem of color difference detection on polycrystalline silicon cells under complex color and texture characteristics, a new method based on support vector machine classification strategy is proposed to detect the color difference of polycrystalline silicon cells. Firstly, color model conversion and channel separation are performed on the pre-processed cell images. The Otsu method is used to perform threshold segmentation processing on the single-channel image, and the region contrast of each threshold image is calculated, and then an appropriate threshold image is selected according to the regional contrast condition. The image features are extracted by the information provided by the threshold image. Finally, the support vector machine classifier is used to determine whether the cell has a color difference defect. The experimental results show that the proposed color difference detection algorithm can achieve high-efficiency detection of color difference defect, and the accuracy, false detection rate and detection time reach 96.88%, 5% and 109 ms.