matlab连通域一遍扫描,高效的一遍扫描式连通区域标记算法.pdf

本文介绍了一种基于Suzuki算法的改进版连通区域标记算法,该算法通过一次扫描即可高效标记二值图像的连通组件,避免了多次遍历,显著提高了标记速度。实验表明,新算法比原始算法快约2倍,且优于近期的一些同类算法。
摘要由CSDN通过智能技术生成

高效的一遍扫描式连通区域标记算法.pdf

Computer Engineering and Applications 计算机工程与应用 2014,50(23) 31

高效的一遍扫描式连通区域标记算法

冯海文 1,2,牛连强 1,刘晓明 2

FENG Haiwen ,2, NIU Lianqiang , LIU Xiaoming2

1.沈阳工业大学 软件学院,沈阳 110023

2.沈阳工业大学 电气工程学院,沈阳 110023

1.School of Software, Shenyang University of Technology, Shenyang 110023, China

2.School of Electrical Engineering, Shenyang University of Technology, Shenyang 110023, China

FENG Haiwen, NIU Lianqiang, LIU Xiaoming. Efficient one-scan algorithm for labeling connected component.

Computer Engineering and Applications, 2014, 50(23):31-35.

Abstract:How to label connected components of a binary image is a basic problem in image processing field. To improve

efficiency, a fast one-scan algorithm to label connected components is presented in this paper on the base of the multiple-

scans algorithm proposed by Suzuki et al. The algorithm runs a forward scan to the object image only once. The node with

the minimum label in the mask of the object pixel is calculated. The node with smaller label is searched in the connected

component by an iterative process, and the connected branch including the note to be updated is linked to it. This tech-

nique can guarantee no loss to the equivalent information. At the same time, the provisional labels in iterative search path

are updated by the minimum label in order to decrease the depth of the branch. The final labels of all nodes are obtained

by scanning the connected table. Dynamic data structure and recursive procedure are not needed in this algorithm, and only

less memory is required. Experiments and analysis show that the algorithm is about 2 times faster than the original one,

and is also faster than some run algorithms proposed recently.

Key words:connected component; labeling algorithm; one-scan; label; binary image; label connected table

摘 要:二值图像的连通区域标记算法是图像处理的一个基本问题。为了提高算法的效率,以 Suzuki 等人提出的多

遍扫描算法为基础,提出

评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值