关键词: 分割, Live-wire算法, 平均代价路径, 带权重的Canny边缘

Abstract: In the original live-wire algorithm, optimality is defined as the minimum cumulative cost path from a seed point to another, where the cumulative cost of a path is the sum of the local edge costs on the path. Consequently, when delineating a boundary with acute concaves and convexes, more seed points are required which can become very time consuming to calculate. This paper proposes an improved live-wire algorithm where the optimality is redefined as the minimum average cost path between two seed points. In theory, the original live-wire algorithm is only a special case of the improved algorithm. Experiments conduct on a variety of image types have shown that this improved live-wire algorithm requires less seed points than the original algorithm when delineating the same boundary and as a result, reduces the time required to complete the calculation.

Key words: segmentation, live-wire, average cost path, weighted Canny edge