改进的分水岭T-Snake轮轨接触区域图像分割算法

IMAGE SEGMENTATION ALGORITHM OF IMPROVED WATERSHED T-SNAKE WHEEL-RAIL CONTACT AREA

Snake算法为一种经典的主动轮廓线模型分割算法,在图像分割领域得到广泛应用。但其存在依赖初始轮廓线的选取和难以提取图像凹陷边缘等缺点。采用结合图像信息熵和高低帽变换的组合标记分水岭算法提取初始边缘轮廓线作为T-Snake的初始轮廓线,并通过引入区域能量项来优化拓扑自适应主动轮廓T-Snake算法。改进后的T-Snake算法能够较好地根据踏面边缘特点自发改变其捕捉方向,对具有凹陷边缘的图像处理能力较强。实验结果证明,该模型应用于列车轮轨接触区域踏面边缘曲线提取中取得了良好的效果。

The Snake algorithm is a classical active contour model segmentation algorithm,which has been widely used in the field of image segmentation.However,it has some disadvantages,such as the dependence on the selection of the initial contour and the difficulty in extracting the concave edges of the image.In this paper,a combined marked watershed algorithm combining image information entropy and high-low hat transformation was used to extract the initial edge contour as the initial contour of T-Snake.The region energy term was introduced to optimize the topology adaptive active contour T-Snake algorithm.The improved T-Snake algorithm was able to change its capture direction spontaneously according to the characteristics of tread edge,and had a strong ability to process images with concave edges.The experimental results show that our model is applied to the extraction of tread edge curve in the train wheel-rail contact area,and has achieved good results.