复杂结构构造区遥感图像的地质信息提取方法——基于小波变换的多层次图像分割

基金项目:

国家自然科学基金项目 49972075

西北大学校内基金项目 00NW32

Geological Information Extracting from Remote Sensing Image in Complex Area: Based on Wavelet Analysis for Automatic Image Segmentation

摘要: 针对基岩裸露的造山带变形强烈、地层组合复杂、解译难度大的特点, 提出了一种有效结合纹理信息和颜色信息的小波帧包变换的多层次图像分割方法, 克服了以往图像处理方法将低层特征孤立起来的缺陷, 在南天山东段铜花山—硫磺山地区地表地质结构的遥感影像解译中获得了令人满意的效果, 为强烈变形的复杂结构构造区从遥感图像上提取地质信息提供了有效的途径
- 小波帧包 /
- 地质信息 /
- 图像分割
Abstract: It is very difficult to interpret remote sensing images in an orogenic belt with complex strata combination and strongly deformed structure. This paper proposes a novel wavelet frame packet analysis approach for multivariable image segmentation. The approach effectively combines both the texture and color information. It has produced good performance in the extraction of geological information in the deformable area in Tonghuashan-Liuhuangshan in the eastern part of Tianshan mountains.
- wavelet packet /
- geological information /
- image segmentation

图 1 小波帧包分解示意

A.小波分解的低频部分(粗糙像); H、V、D.分别表示水平方向、垂直方向、对角线方向的高频部分(细节)

Fig. 1. Tree structure of wavelet packet analysis

下载: 全尺寸图片幻灯片

图 2 原始遥感图像(TM4)

Fig. 2. Original image of remote sensing (TM4)

下载: 全尺寸图片幻灯片

图 3 AA点宏观分类

Fig. 3. Rude classification of AA point

下载: 全尺寸图片幻灯片

图 4 DA点细节分类

Fig. 4. Classification of DA point in details

下载: 全尺寸图片幻灯片

图 5 铜花山—硫磺山地区遥感影像综合解译

Fig. 5. Interpreter image of Tonghuashan-Liuhuangshan area

下载: 全尺寸图片幻灯片

[1]	杨觊, 卢健, 林开愚. 遥感图像处理原理和方法[M]. 北京: 测绘出版社, 1988. Yang K, Lu J, Lin K Y. The principle and methods of remote sensing image processing[ M]. Beijing: Servey and Mapping Press, 1988.
[2]	朱亮璞. 遥感地质学[M]. 北京: 地质出版社, 1994. Zhu L P. Remote sensing geology[M]. Beijing: Geological Publishing House, 1994.
[3]	崔锦泰. 小波分析导论[M]. 程正兴, 译. 西安: 西安交通大学出版社, 1995. Cui J T. Introduction of wavelet analysis[M]. Translation by Cheng Z X. Xi'an: Xi'an Transportation University Press, 1995.
[4]	Smith J R, Chang S F. Transform features for texture classification and discrimination in large image database[J]. ICIP, 1994, 3: 307-411.
[5]	Unser M. Texture classification and segmentation using wavelet frame[ J]. IEEE Trans Image Processing, 1995, 4(11): 1549-1560.
[6]	Liu J F, MIEE J C. An efficient and effective texture classification approach: using a new notion in wavelet theory[J]. ICPR, 1996, 2: 820-824.