| 基于不变矩的高分辨率遥感图像建筑物提取方法 |
| |
| 引用本文： | 高 薇,王 珂,原发杰,余 涛,李玉霞,高 斌.基于不变矩的高分辨率遥感图像建筑物提取方法[J].计算机应用研究,2014,31(2):622-624. |
| |
| 作者姓名： | 高 薇 王 珂 原发杰 余 涛 李玉霞 高 斌 |
| |
| 作者单位： | 1. 电子科技大学 自动化工程学院, 成都 611731; 2. 中国科学院遥感与数字地球研究所, 北京 100101 |
| |
| 基金项目： | 民用航天“十二五”预研项目（Y1K00200KJ）; 中国科学院对外重点合作项目（GJHZ1003） |
| |
| 摘 要： | 为了有效地对图像进行特征提取, 利用不变矩算法对IKONOS和WorldView两种高分辨率遥感图像的城市建筑物地区进行提取。首先将图像数据经过Canny边缘检测和标记分水岭分割, 然后在此基础上分别利用胡氏不变矩和仿射不变矩对图像进行特征提取; 最后通过实验结果的评价可以证明在建筑物的特征提取上, 仿射不变矩比胡氏不变矩的提取效果更加显著, 进而也证明了利用不变矩算法对高分辨率遥感图像建筑物特征提取这一方法是可行且有效的。
|
| 关 键 词： | 高分辨率遥感图像 边缘检测 不变矩 特征提取 |
Building information extraction of high resolution remote sensing image based on moment invariants |
| |
| GAO Wei,WANG Ke,YUAN Fa-jie,YU Tao,LI Yu-xi,GAO Bin.Building information extraction of high resolution remote sensing image based on moment invariants[J].Application Research of Computers,2014,31(2):622-624. |
| |
| Authors: | GAO Wei WANG Ke YUAN Fa-jie YU Tao LI Yu-xi GAO Bin |
| |
| Affiliation: | 1. School of Automation Engineering, University of Electronic Science & Technology of China, Chengdu 611731, China; 2. Institute of Remote Sensing & Digital Earth, Chinese Academy of Sciences, Beijing 100101, China |
| |
| Abstract: | In order to effectively extract features from remote sensing images, this paper utilized moment invariants algorithm to extract the urban building area from high-resolution remote sensing images IKONOS and WorldView. It firstly processed the image data by Canny algorithm and watershed segmentation. Then, it used Hu moment invariants and affine moment invariants for feature extraction. The experiment results show that affine invariant moments has a better effect than Hu moment invariants in feature extraction of buildings, and that the use of moment invariants algorithm in building feature extraction for high-resolution remote sensing data is feasible. |
| |
| Keywords: | high-resolution remote sensing images edge detection moment invariant feature extraction |
|
| 点击此处可从《计算机应用研究》浏览原始摘要信息 |
|
点击此处可从《计算机应用研究》下载全文 |
|
