液晶与显示, 2014, 29 (4): 592, 网络出版: 2014-06-10   

基于图论阈值算法的图像分割研究

Image mosaic research based on wavelet and rough set algorithm
作者单位
黄淮学院 信息工程学院, 河南 驻马店 463000
摘要
为了提高阈值分割图像的质量, 提出了采用图论阈值算法。首先, 构造图论和图像的映射函数关系, 每个顶点通过点来映射, 每条边通过线来映射。用基于区域属性的图像边缘决策表, 不同像素点或不同组像素点之间的灰度特征差作为权重系数, 通过基于决策属性权重来构造像素联系图; 然后, 采用聚类法计算像素到目标类和背景类的相似程度, 最小生成树策略解决伪割集问题; 最后, 给出图像阈值设定以及算法流程。实验仿真表明, 本文算法的分割图像效果清晰, 消除了图像分割中存在的过合并和欠合并现象, 本文算法的信息熵为28.780 3 bit,处理时间为1.454 3 s。满足分割结果中对执行时间少、信息含量大等要求。
Abstract
In order to improve the quality of image segmentation, Graph Theory Threshold Algorithm is established. First, map function is constructed between the graph and image, each vertex is mapped by point, each edge is mapped by line. Second, decision table of image edge is set by attribute of area, gray feature is defined weight coefficient between pixel of different points and groups, decisions of attribute weight are constructed with figure of pixel. Third, similar of pixel is computed by cluster between object and background class, minimum spanning tree method solved the problem of pseudo cut set. Finally, process is described. The experiments show this algorithm made image edge clearly, eliminate the over and under of merger phenomenon, entropy of information is 28.7803 bit, time are 1.454 3 s. It takes less time and has large information.

张健, 李宏升. 基于图论阈值算法的图像分割研究[J]. 液晶与显示, 2014, 29(4): 592. ZHANG Jian, LI Hong-sheng. Image mosaic research based on wavelet and rough set algorithm[J]. Chinese Journal of Liquid Crystals and Displays, 2014, 29(4): 592.

本文已被 5 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!