Image Analysis by Discrete Relative Vector Field

被引：0
|
作者：
Zhuang, Xiaodong [1 ]
Mastorakis, Nikos E. [2 ]
Kaminaris, Stavros D. [3 ]
Chi, Jieru [1 ]
Wang, Hanping [1 ]
机构：
[1] Qingdao Univ, Elect Engn Dept, Elect Informat Coll, Qingdao 266071, Peoples R China
[2] Hellen Naval Acad, Dept Elect Engn & Comp Sci, Piraeus, Greece
[3] Piraeus Univ Appl Sci, Dept Elect Engn, Athens, Greece
来源：
2017 FOURTH INTERNATIONAL CONFERENCE ON MATHEMATICS AND COMPUTERS IN SCIENCES AND IN INDUSTRY (MCSI) | 2017年
关键词：
image transform; relative vector field; physics inspired; image structure; image segmentation; REPRESENTATION;
D O I：
10.1109/MCSI.2017.15
中图分类号：
TP39 [计算机的应用];
学科分类号：
081203 ; 0835 ;
摘要：
A novel method of relative vector field is proposed for gray-scale image analysis by imitating the electro-static field. It extends the mathematical form of electro-static field by introducing the item of gray-scale difference, which represents the local difference between adjacent image points. The local image regions can be segmented based on the relative vector field with a proposed region-growing method, and practical image segmentation can be archived by merging primitive regions. The experimental results prove the effectiveness of the proposed method.
引用
收藏
页码：30 / 36
页数：7
相关论文
共 50 条
12345