STUDY OF IMAGE REGISTRATION APPROACHES ON POINT CLOUD DATASETS

To identify the object in any view, Image Registration process plays an important role. Image Registration process is an alignment process of two images, which is called ‘Reference images’ and ‘Input images’ that images can be taken of same scene but from different viewpoints, different sensors or from different times. In image processing, there are different types of approaches for image registration, and that approaches also categorized in differentdifferent parts accordingly their features. [1] Image Registration itself includes methods of registration which are useful to detect or match the image with the reference image. Nowadays, Image registration strategies are applied mostly in medical, satellites, weather-forecasting, remote sensing, signal processing field. In this Review paper we study on the structure of image registration methods as well as the techniques which going to be used. We believe that Image registration will be useful research area for the researchers to implement alternate approaches of image registration.