Abstract: For the surface crack detection of the low voltage current transformer, the algorithm based on percolation algorithm and OPTA（One-Pass Thinning Algorithm） is proposed. First, the gray image of the transformer is acquired. Second, from the change of pixel value and brightness of the crack region, binarization?image can be obtained by setting the threshold and the seed pixels. Third, using the connectivity of crack, the skeleton of ROI which consists of a single pixel is obtained through OPTA. Finally, accoding to the characteristic of bifurcation, the number of pixels in the neighborhood being more than two will be judged as crack. Experimental results show that the ROI can be obtained from the image effectively, and its linear features are maintained. The improved OPTA makes the skeleton of ROI a single pixel image and the accuracy of proposed method is above 97%.

Key words:  percolation algorithm, improved One-Pass Thinning Algorithm（OPTA）, skeleton

摘要： 针对低压电流互感器表面裂纹的提取与判定，提出基于渗透算法和改进型OPTA（One-Pass Thinning Algorithm）的互感器表面裂纹检测算法。首先获取互感器表面的灰度图像；其次根据裂纹区域像素值、亮度变化，通过设定种子像素点、亮度阈值，利用渗透算法渗透得到二值图；再次从裂纹连通性入手，利用改进型OPTA提取ROI（Region of Interest）的骨架，骨架由单像素点组成；最后利用裂纹具有分叉性的特点，像素点的邻域点个数超过2的即可判定为裂纹。实验表明，渗透算法能够有效地从图像中提取出ROI，并保持了ROI的线性特征，改进型OPTA使ROI完全细化为单像素图像，提出的邻域点判别法检测效率在97%以上，相较于所提其他检测方法有明显提高。

关键词: 渗透算法, 改进型OPTA, 骨架