基于Harris-Laplace特征点的多时相SPOT图像配准方法  

Multi-temporal SPOT image registration based Harris-Laplace feature points

作　　者：杨朝辉[1] 

机构地区：[1]苏州科技学院环境科学与工程学院,江苏苏州215011

出　　处：《苏州科技学院学报（自然科学版）》2012年第2期72-75,共4页Journal of Suzhou University of Science and Technology （Natural Science Edition）

基　　金：江苏省高校自然科学基金资助项目(10KJB42000)

摘　　要：提出一种改进的多时相SPOT图像配准方法。首先通过分块提取H-L特征点,确保特征点分布均匀;然后对H-L特征点进行匹配,通过RANSAC一致性检验构建配准模型;最后利用图像重采样得到图像配准结果。实验结果表明,该方法能够有效地解决不同时相SPOT图像在几何变形、地物改变等情况下的配准,具有较高的自动化程度与实用性。A new method for multi-temporal SPOT image registration is proposed. Firstly, ensuring the feature points averagely distributed over the image, Harris-Laplace points were extracted from sub-image. Secondly, H-L feature points were registered. The registered model was examined by random sample consensus （RANSAC）. Finally, image registration was achieved by resampling the image. Experimental results show that this method can effectively have image registration of multi-temporal SPOT images under the transformation of geometry and surface features, and that it takes high automation and practicability.

关 键 词：SPOT图像 多时相 Harris—Laplace特征点 图像配准 随机采样一致性 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]