摘要
Abstract
The spectral character is the basic character in remote sensing images, and is naturally being used the most widely in remote sensing image retrieval. Considering the characteristic of multi-spectral or hyper-spectral remote sensing images, K-L transformation based image retrieval method is proposed in this paper. Instead of original multi-spectral image, image retrieval can be carried out using the histogram of the first three principal component images; the distance of the remote sensing images can be calculated after extracting the spectral character. This is also proved to be effective and time saving by the experiments in this paper.关键词
摄影测量与遥感/遥感图像检索/光谱特征/多光谱图像/K-L变换Key words
photogrammetry and remote sensing/remote sensing image retrieval/spectral character/multi-spectral images/K-L transformation分类
天文与地球科学
