计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (13): 197-198.DOI: 10.3778/j.issn.1002-8331.2010.13.059

• 图形、图像、模式识别 • 上一篇    下一篇

使用Bhattacharyya系数的道路自动提取

刘朋飞1,张剑清1,2,潘 励1,刘永亮3   

  1. 1.武汉大学 遥感信息工程学院,武汉 430079
    2.武汉大学 测绘遥感信息工程国家重点实验室,武汉 430079
    3.四川中水成勘院测绘工程有限责任公司,成都 610000
  • 收稿日期:2010-03-01 修回日期:2010-04-16 出版日期:2010-05-01 发布日期:2010-05-01
  • 通讯作者: 刘朋飞

Automatic road extraction with Bhattacharyya coefficient

LIU Peng-fei1,ZHANG Jian-qing1,2,PAN Li1,LIU Yong-liang3   

  1. 1.School of Remote Sensing and Information Engineering,Wuhan University,Wuhan 430079,China
    2.State Key Lab of Information Engineering in Surveying,Mapping and Remote Sensing,Wuhan University,Wuhan 430079,China
    3.Chengdu Hydroelectric Investigation & Design Institute of SPC,Chengdu 610000,China
  • Received:2010-03-01 Revised:2010-04-16 Online:2010-05-01 Published:2010-05-01
  • Contact: LIU Peng-fei

摘要: 介绍了基于Bhattacharyya系数的小比例尺正射影像道路提取方法。首先假设道路点的灰度特征符合标准正态分布,建立理想道路模型,在影像中通过计算Bhattacharyya系数评价该位置为道路点的可能性,系数较大者为道路点。实验证明该方法具有一定的可行性。

关键词: 道路中心点, 正态分布, Bhattacharyya系数, 理想模型

Abstract: In this paper,a method on road extraction from small scale orthoimage based on Bhattacharyya coefficient is introduced.First,the road is assumed as the set of the center point whose ideal gray distribution feature can be supposed as normal distribution(also known as Gauss distribution),and then Bhattacharyya coefficient is calculated for each pixel of the candidate image.And the ones with greater coefficient values can be seen as road center points.Experiments prove the validity of this algorithm.

Key words: road center point, normal distribution, Bhattacharyya coefficient, ideal model

中图分类号: