基于近似最小距离场的二维图像骨架提取方法  

2D image skeleton generation based on approximate minimum distance field

作　　者：庄彩云[1] 熊平[1] 

机构地区：[1]中南大学地球科学与信息物理学院,长沙410083

出　　处：《计算机工程与应用》2013年第21期164-167,共4页Computer Engineering and Applications

基　　金：国家自然科学基金(No.30371626)

摘　　要：提出了基于近似最小距离场提取二值图像的8-连通骨架的算法。该算法对图像中的每个像素根据其与边界的相对距离进行整数编码,形成近似最小距离场,将该距离场中的几何邻接的、具有局部最大值的像素形成聚类,对聚类进行细化,用最短路径将不同的细化后的聚类连接起来。该算法简单,将其在实验数据集上进行实验,结果证明算法具有很高的效率。This paper proposes an algorithm for extracting 8-connected skeletons of 2D binary images. Each interior pixel in the 2D image is encoded with an integer code according to its relative distance from the object border to form an approximate minimum distance field. Cluster is defined as a set of geometrically connected local maximum pixels with the same distance value. And all the clusters are thinned, and connected with the shortest paths. The proposed algorithm is simple, and the results acquired by the algorithm on an experimental data demonstrate its efficiency.

关 键 词：近似最小距离场 2D二值图像 像素编码 聚类 最短路径 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]