摘要
针对多晶硅太阳电池片裂纹检测中背景干扰和对比度较低等检测难点,提出基于微分匹配均值和匹配滤波器的裂纹检测算法。首先用基于高斯函数的匹配滤波器从多角度、多方向提取裂纹空间特征的算子,增强图像;同时用微分匹配均值滤波器区分裂纹和晶粒结构,有效提取裂纹;然后根据阈值公式筛选裂纹结构;最后用Hough变换去除栅线,并采用形态学方法提取出完整裂纹结构。实验结果表明,提出的裂纹检测算法能够精确、有效地检测出多晶硅太阳电池片的裂纹,解决了紊乱的晶粒背景干扰检测的问题,相比其他几种算法兼具鲁棒性和快速性。
Aiming at the detection difficulties caused by background interference and low contrast in the crack detection of polysilicon solar cells,a crack detection algorithm based on differential matched mean and matched filters is put forward.Firstly,a matched filter based on Gaussian function is used to extract the spatial characteristics of cracks from multiple angles and directions to enhance the image.At the same time,a differential matched mean filter is used to distinguish cracks and grain structures to effectively extract cracks.Then,crack structures are filtered according to the threshold formula.Finally,Hough transform is used to remove the grid lines,and the complete crack structure is extracted by morphological methods.Experimental results show that the crack detection scheme proposed in this paper can accurately and effectively detect the cracks of polycrystalline silicon solar cells,solve the problem of disordered grain background interference detection,and it is better than several other algorithms both in robustness and speed.
作者
张玉婷
郑茜颖
俞金玲
ZHANG Yuting;ZHENG Qianying;YU Jinling(College of Physics and Information Engineering,Fuzhou University,Fuzhou Fujian 350000,China)
出处
《电源技术》
CAS
北大核心
2021年第11期1486-1489,共4页
Chinese Journal of Power Sources
基金
国家自然科学基金资助项目(61471124)
福建省科技厅工业引导性重点项目(2020H0007)。
