基于L型结构中心线的SAR图像建筑物提取方法
针对在高分辨率SAR图像上具有明显L型结构高亮特征的建筑物目标,提出了一种提取高分辨率SAR图像建筑物L型结构中心线,并进而提取建筑物几何信息的方法。运用基于Gabor纹理特征和模糊C均值的方法对SAR图像进行分割,再结合骨架提取、骨架跟踪、最小外接矩形提取、最小二乘准则等技术实现了L型结构中心线的提取,最后利用中心线获取了建筑物的长度、宽度和方位角信息。基于机载SAR图像的实验表明,利用提出的方法从SAR图像提取的建筑物几何结构和方位信息具有较高的精度。...
Saved in:
Published in | 计算机应用研究 Vol. 32; no. 6; pp. 1888 - 1891 |
---|---|
Main Author | |
Format | Journal Article |
Language | Chinese |
Published |
中国科学院大学,北京100049%中国科学院遥感与数字地球研究所,北京,100101%中国科学院电子学研究所,北京,100190
2015
中国科学院遥感与数字地球研究所,北京100101 |
Subjects | |
Online Access | Get full text |
ISSN | 1001-3695 |
DOI | 10.3969/j.issn.1001-3695.2015.06.066 |
Cover
Summary: | 针对在高分辨率SAR图像上具有明显L型结构高亮特征的建筑物目标,提出了一种提取高分辨率SAR图像建筑物L型结构中心线,并进而提取建筑物几何信息的方法。运用基于Gabor纹理特征和模糊C均值的方法对SAR图像进行分割,再结合骨架提取、骨架跟踪、最小外接矩形提取、最小二乘准则等技术实现了L型结构中心线的提取,最后利用中心线获取了建筑物的长度、宽度和方位角信息。基于机载SAR图像的实验表明,利用提出的方法从SAR图像提取的建筑物几何结构和方位信息具有较高的精度。 |
---|---|
Bibliography: | 51-1196/TP high resolution SAR image; building; L-shaped structure; center line; extraction Nil Xu, Zhang Fengli, Wang Guojun, Zhou Liangjiang, Shao Yun, Liang Xingdong (1. Institute of Remote Sensing & Digital Earth, Chinese Academy of Sciences, Beijing 100101, China; 2. University of Chinese Academy of Sciences, Beijing 100049, China; 3. Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China) Aiming at building of L-shapes structure in high resolution SAR image, this paper proposed an approach to extract center line of building L-shaped structure, and retrieved geometrical information of building by center line. The proposed ap- proach used Gabor texture feature and fuzzy C-means to fulfill building segmentation. Then it used skeleton extraction, skeleton tracking, minimum bounding rectangle extraction, and least squares to achieve the extraction of center line. It achieved length, width, azimuth of building by center line. The experiment based on airborne SAR image shows the approach can |
ISSN: | 1001-3695 |
DOI: | 10.3969/j.issn.1001-3695.2015.06.066 |