硅片缺陷检测中的图像分割方法
|
摘要
在利用机器视觉系统检测太阳能硅片外观缺陷的过程中,分割阈值的选择一直是一个较大的挑战。针对此问题提出了一种基于局部灰度特征的自适应阈值选择方法。首先,对初始图像进行亮度校正和调整,解决外部光源造成的亮度分布不均问题;其次,在不同时间和不同区域采集图像块,分析它们的灰度差异,以此估算环境光线的动态变化;最后,在人工设定目标阈值的基础上,分割阈值作限定性的动态调整。实验结果表明,该方法不仅具有对环境光线鲁棒性的优点,而且可以避免现存的算法缺乏目标性的缺点,因此具有较好的可行性。
In the process of detecting the appearance defects of solar wafers by a machine vision system,the selection of an appropriate threshold has been a great challenge.In order to solve this problem,an adaptive threshold selection method based on local gray feature was proposed.Firstly,luminance values were corrected in original images to avoid uneven brightness distribution caused by external light sources.Secondly,image blocks were sampled at different times and in different areas.The differnces of their gray values were analyzed to estimate the changes of the external environment light.Finally,the dynamic adjustment of the threshold was defined based on the manual setting of the target threshold.Experimental results show that the proposed method not only has the advantages of robustness to the environment,but also can avoid the disadvantages of lack of targets in the existing algorithms.Therefore,the method has good feasibility.
In the process of detecting the appearance defects of solar wafers by a machine vision system,the selection of an appropriate threshold has been a great challenge.In order to solve this problem,an adaptive threshold selection method based on local gray feature was proposed.Firstly,luminance values were corrected in original images to avoid uneven brightness distribution caused by external light sources.Secondly,image blocks were sampled at different times and in different areas.The differnces of their gray values were analyzed to estimate the changes of the external environment light.Finally,the dynamic adjustment of the threshold was defined based on the manual setting of the target threshold.Experimental results show that the proposed method not only has the advantages of robustness to the environment,but also can avoid the disadvantages of lack of targets in the existing algorithms.Therefore,the method has good feasibility.
引文
[1]张舞杰,李迪,叶峰.硅太阳能电池视觉检测方法研究[J].计算机应用,2010,30(1):249-252.
[2]LI W-C,TSAI D-M.Automatic saw-mark detection in multicrystalline solar wafer images[J].Solar Energy Materials and Solar Cells,2011,95(8):2206-2220.
[3]TSAI D M,LUO J Y.Mean shift-based defect detection in multicrystalline solar wafer surfaces[J].IEEE Transactions on Industrial Informatics,2011,7(1):125-135.
[4]梁华为.直接从双峰值方图确定二值化阈值[J].模式识别与人工智能,2002,15(2):253-256.
[5]杨莉,杨新.基于区域划分的曲线演化多目标分割[J].计算机学报,2004,27(3):420-425.
[6]王斌,李洁,高新波.一种基于边缘与区域信息的先验水平集图像分割方法[J].计算机学报,2012,35(5):1067.
[7]CANNY J.A computational approach to edge detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1986,8(6):679-698.
[8]刘锁兰,杨静宇.基于模糊理论的2维隶属划分Renyi熵分割算法[J].中国图象图形学报,2009,14(2):323-327.
[9]21 N.A threshold selection method from gray-level histograms[J].IEEE Transactions on Systems,Man and Cybernetics,1979,9(1):62-66.
[10]张翰进,傅志中,念蓓,等.双峰法与otsu法结合在太阳能电池缺陷检测中的应用[J].计算机系统应用,2012,21(1):115-117.
[11]葛琦,韦志辉,肖亮,等.基于局部特征的自适应快速图像分割模型[J].计算机研究与发展,2015,50(4):815-822.
[12]郝颖明,朱枫.2维Otsu自适应阈值的快速算法[J].中国图象图形学报,2005,10(4):484-488.
[13]李哲学,陈树越.快速多阈值图像分割法[J].计算机应用,2010,30(5):1335-1337.
[2]LI W-C,TSAI D-M.Automatic saw-mark detection in multicrystalline solar wafer images[J].Solar Energy Materials and Solar Cells,2011,95(8):2206-2220.
[3]TSAI D M,LUO J Y.Mean shift-based defect detection in multicrystalline solar wafer surfaces[J].IEEE Transactions on Industrial Informatics,2011,7(1):125-135.
[4]梁华为.直接从双峰值方图确定二值化阈值[J].模式识别与人工智能,2002,15(2):253-256.
[5]杨莉,杨新.基于区域划分的曲线演化多目标分割[J].计算机学报,2004,27(3):420-425.
[6]王斌,李洁,高新波.一种基于边缘与区域信息的先验水平集图像分割方法[J].计算机学报,2012,35(5):1067.
[7]CANNY J.A computational approach to edge detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1986,8(6):679-698.
[8]刘锁兰,杨静宇.基于模糊理论的2维隶属划分Renyi熵分割算法[J].中国图象图形学报,2009,14(2):323-327.
[9]21 N.A threshold selection method from gray-level histograms[J].IEEE Transactions on Systems,Man and Cybernetics,1979,9(1):62-66.
[10]张翰进,傅志中,念蓓,等.双峰法与otsu法结合在太阳能电池缺陷检测中的应用[J].计算机系统应用,2012,21(1):115-117.
[11]葛琦,韦志辉,肖亮,等.基于局部特征的自适应快速图像分割模型[J].计算机研究与发展,2015,50(4):815-822.
[12]郝颖明,朱枫.2维Otsu自适应阈值的快速算法[J].中国图象图形学报,2005,10(4):484-488.
[13]李哲学,陈树越.快速多阈值图像分割法[J].计算机应用,2010,30(5):1335-1337.