一种改进的图像骨架提取算法  

An Improved Image Skeleton Extraction Algorithm

作　　者：叶福玲[1,2] YE Fu-ling1,2(1.College of Mathematics and Computer Science; 2.National Experimental Teaching Demonstrating Center of Network and Information Security and Computer Technology, Fuzhou University, Fuzhou 350116,China)

机构地区：[1]福州大学数学与计算机科学学院,福州350116 [2]福州大学网络信息安全与计算机技术国家级实验教学示范中心,福州350116

出　　处：《西昌学院学报（自然科学版）》2018年第3期91-93,123,共4页Journal of Xichang University(Natural Science Edition)

摘　　要：针对Zhang细化算法的不足提出改进算法。首先,根据提取的骨架特点设计细化模板,将骨架进行细化并且保持骨架的连通性和满足单像素,便于后续处理;其次,利用像素找出骨架的分支点,骨架减去分支点形成多个不连通区域,根据骨架特点选择不同的阈值,去除小于该阈值的连通分支,从而去除骨架毛刺,该步骤高效地除去了多余的噪音和毛刺,优化了骨架的视觉效果。仿真实验结果表明用改进算法提取图像骨架比Zhang细化算法得到的骨架效果更好,不但能够根据阈值的选择来除掉长短不一的毛刺,且不会破坏物体结构中重要的骨架。An improved algorithm is proposed for the shortage of Zhang refinement algorithm. First of all, therefinement template is designed according to the characteristics of the extracted skeleton, and the skeleton is refinedand the skeleton connectivity is maintained and the single pixel is satisfied for subsequent processing. Secondly，using the pixels to find the skeleton of the branch point, the skeleton minus the branch points to form a number ofnon-connected areas. Different thresholds are selected according to the skeleton characteristics, and the connectedbranches which are smaller than the threshold will remove the skeleton glitches. This step effectively removes excessnoise and glitches, optimizing the visual effect of the skeleton. The simulation results show that the improvedskeleton is better than the Zhang refinement algorithm, not only in accordance with the threshold of choice to get ridof the length of the burr, and will not destroy the important structure of the skeleton.

关 键 词：骨架 细化 单像素 去除毛刺 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]