重叠纤类图像的凹点匹配和分割算法

Concave points matching and segmentation algorithm for overlapped fiber image

为解决混纺纤维类图像中多根纤维的分离和识别难题,提出一种基于凹点匹配的新方案来实现重叠纤维分割。在图像轮廓寻找角点,利用三角形矢量面积法找到角点中包含的凹点;通过凹点、该凹点的前继点(或后继点)和目标凹点构成的三角形,结合凹点的几何特征来判断该凹点和目标凹点是否匹配;通过将匹配凹点连线来实现图像中的纤维分割。与现有分割算法相比,新方法采用了凹点距离和三角形构造原理相结合的机制来实现匹配。结果表明,该算法可适用于多根纤维相互黏连或交叉的复杂场景,且有较高的分割精度。

In order to identify the different components which are adhered or crossed each other in a hybrid fiber image,a fiber segmentation algorithm based on concave points matching in the overlap area was proposed. The detection of contour corners was firstly obtained,and the vector triangle area method was also used to find contour concave points from the corner set. For a given concave point,the precursor point（ or subsequent point） and the target point were selected to form a triangle. The triangle and other significant features were used to judge whether the concave point and target point were matched. At last,by the connection of the matching concave points,the segmentation of overlapped fiber image could be realized. Compared with the existing algorithms of segmentation,the novel method employed a mechanism combined with distance between concave points and triangular construction principle to achieve the purpose of matching. The experimental results show that the algorithm can deal with the situation of adhesive fiber and cross fiber,with high segmentation accuracy over 80% in complicated scenes.