Indian Journal of Science and Technology

Abstract

Background: Image registration is one of the challenging issues in remote sensing applications. In last few years a wide variety of image registration techniques have been proposed. This paper reviews various registration techniques of satellite images. The approach discussed in this paper includes feature extraction and selection, feature matching, transform model estimation, transformation and resampling. Methods: The methods of registrations which are applicable to multi temporal, multi spectral, multi angle and multisensory are discussed. Automatic registration of remote sensing images deals with various problems such as intensity variations, illumination changes, geometric distortions, noise etc. For solving all the registration problems various techniques are presented for various applications. There are two types of registration involved, one type is intensity based and other one is feature based. Literature evaluation of various techniques describes the methods applicable for each application. Results: The findings shown that SIFT is the best suitable method for feature extraction. This survey helps researchers for attain information about registration of remote sensing images. Application: Image registration can be applied in various applications such as Multispectral classification, Environmental monitoring, Change detection, Image mosaicking, Weather forecasting, creating super-resolution images etc.
Keywords: Feature Extraction, Feature Matching, Image Registration, Image Resampling, Remote Sensing, Transformation Model