Image registration is very important part in image processing, in which two or more images are align in common coordinate system which are taken from different viewpoint, at different times, depths and by different sensor. Main purpose of this paper is to give detail conceptual information of important existing concepts of image registration. Also the image registration process is discussed briefly as per its stages i.e. feature detection, feature matching, transform model estimation, image resampling and transformation. Limitations of some existing methods are mentioned in this paper.
Main aim of this paper is to provide information on Image registration methods. Image Registration is the process of aligning two images which can be taken at different time, view point and sensors. It aligns the two images of same scene (reference image and sense image) into a common coordinate system. We have to align them in order to monitor subtle changes between the two. We can further divide the image registration in four parts: feature detection, feature mapping, mapping function, transformation and resampling. The image registration can be of two types. One is area based and other is feature based. The purpose of this paper is to provide a comprehensive review of the existing literature available on Image registration methods so that it can be helpful to further study the new things in image registration. This can be helpful to the new researchers to implement such methods. We can apply this method into real life.
Image registration, transformation, resampling
