作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2012, Vol. 38 ›› Issue (12): 191-193. doi: 10.3969/j.issn.1000-3428.2012.12.057

• 图形图像处理 • 上一篇    下一篇

基于水平集的SAR图像建筑物区域检测

王瑞花 1,宋建社 1,高 江 2,王晓梅 2   

  1. (1. 西安高技术研究所,西安 710025;2. 第二炮兵青州士官学校,山东 青州 262500)
  • 收稿日期:2011-07-18 出版日期:2012-06-20 发布日期:2012-06-20
  • 作者简介:王瑞花(1983-),女,博士研究生,主研方向:SAR信息处理;宋建社,教授、博士生导师;高 江,助教;王晓梅, 讲师
  • 基金资助:
    国家自然科学基金资助项目(61072141)

Building Areas Detection in SAR Images Based on Level Set

WANG Rui-hua 1, SONG Jian-she 1, GAO Jiang 2, WANG Xiao-mei 2   

  1. (1. Xi’an Research Institute of High-tech, Xi’an 710025, China; 2. The Second Artillery Sergeant College, Qingzhou 262500, China)
  • Received:2011-07-18 Online:2012-06-20 Published:2012-06-20

摘要: 根据合成孔径雷达(SAR)图像建筑物成像的特点,采用基于水平集的分割方法对SAR图像建筑物区域进行检测。结合基于指数小波提取的SAR图像梯度边缘信息,以及基于变差函数提取的SAR纹理特征,建立能量函数模型,进行曲线演化分割从而提取目标。实验结果表明,与基于灰度信息的模型相比,该模型可以准确、快速地检测出大场景SAR图像中的建筑物区域。

关键词: 合成孔径雷达, 目标检测, 建筑物, 水平集, 变差函数, 指数小波

Abstract: According to the characteristic of building areas in Synthetic Aperture Radar(SAR) images, a method based on the level set is used to detect building areas and building targets. It uses the boundary feature based on exponential wavelet and texture feature based on variogram to build the energy criterion model, and then the evolutive curve is implemented by minimizing the energy criterion via variational level set approach to detect the building areas. Compared with the traditional energy criterion model based on the gray information, experimental results show that the proposed method is effective and can detect building areas of the large SAR image accurately.

Key words: Synthetic Aperture Radar(SAR), target detection, building, level set, variogram, exponential wavelet

中图分类号: