Title :
A Registration Algorithm Based on Feature Point
Author :
Ma Jianzhe ; Li Lipeng
Author_Institution :
Beijing YuYing Middle Sch., Beijing, China
Abstract :
The existing speckle noise in SAR image produces serious impact on the registration, this paper presents a algorithm of multi-level non-linear weighted mean median filtering, smoothly filtering SAR, effectively suppressing speckle noise in images while maintaining the image geometry, thus the noise on the registration accuracy can be reduced. For registration of high-resolution remote sensing image, the paper improves the traditional method with registration between SAR and panchromatic image, an automatic registration algorithm based on the control point is proposed, which determine the registration evaluation through subjective and objective methods. Experimental results show that the algorithm can select and match control points without human intervention, with fast calculating speed, and it reaches sub-pixel level registration, with good registration effect.
Keywords :
image denoising; image registration; interference suppression; median filters; radar imaging; remote sensing; smoothing methods; synthetic aperture radar; SAR image; control points; image registration; nonlinear weighted mean median filtering; panchromatic image; remote sensing image; smooth filtering; speckle noise suppression; Accuracy; Feature extraction; Filtering; Filtering algorithms; Image registration; Noise; Speckle; Control points; Registration; SAR; Sub-pixel;
Conference_Titel :
Information Technology, Computer Engineering and Management Sciences (ICM), 2011 International Conference on
Conference_Location :
Nanjing, Jiangsu
Print_ISBN :
978-1-4577-1419-1
DOI :
10.1109/ICM.2011.337