会议专题

Multi-Spectral Remote Sensing Image Registration Based on SURF

As remote sensing breaking into all fields of professions, the needs of it are increasing, which give power to its development. However, before using the images, we must process them first. And image registration plays an significant role in remote sensing image processing. Now more and more sensors appeard, multi-spectral remote sensing images are used in widely application all the way. At the same time, for its characteristics of mass data and illumination variation, it is extremely urgent to find a way to register multi-spectral remote sensing images automatically and accurately. SURF is a kind of feature extraction algorithm with scale-invariation, rotation-invariation, iliumination-invariation and speediness. This paper applied SURF on multi-spectral image registration in different platforms, and got satisfied results, which verifies the feasibility of SURF in multi-spectral image registration.

image registration Remote Sensing multi-spectral SURF(Speeded-Up Robust Features)

Yunfei Lu Haimeng Zhao Bo Li Lei Yan

Institute of Remote Sensing & Geographic Information System,Beijing Key Lab of Spatial Information Integration & Its Applications, Peking University,Beijing, China

国际会议

2010 International Conference on Circuit and Signal Processing(2010年电路与信号处理国际会议 ICCSP 2010)

上海

英文

632-635

2010-12-25(万方平台首次上网日期,不代表论文的发表时间)