基于边缘特征匹配的遥感影像变化检测预处理方法

基金项目: 国家973计划资助项目(2006CB701302);全国优秀博士学位论文作者专项资金资助项目(200142);国家教育部长江学者和创新团队发展计划资助项目(IRT0438)。

详细信息

Image Preprocessing Method Based on Edge Feature Matching for Change Detection in Multi-temporal Remotely Sensed Imageries

¹ School of Remote Sensing and Information Engineering, Wuhan University, 129 Luoyu Road, Wuhan 430079, China

Funds: 国家973计划资助项目(2006CB701302);全国优秀博士学位论文作者专项资金资助项目(200142);国家教育部长江学者和创新团队发展计划资助项目(IRT0438)。

摘要: 提出了一种基于边缘特征匹配的遥感影像变化检测预处理方法,在进行不同时相遥感影像配准时,一并解求两期影像的辐射校正系数,同时实现两期影像的配准和辐射校正。实验表明,经预处理后,两期影像的色调基本一致,地物没有明显的几何变形。
- 变化检测 /
- 特征匹配 /
- 配准 /
- 辐射校正 /
- 随机采样一致性原则
Abstract: A novel image preprocessing method is proposed. The method is based on edge feature matching and is designed for change detection to multi-temporal remotely sensed imageries. It can make the image registration and relative radiometric correction at the same time. The experiment results show that the preprocessed images have identical tonality and high registration accuracy. These meet the requirement for multi-temporal imagery change detection.
- image change detection /
- feature-based matching /
- registration /
- radiometric correction /
- RANSAC