Inclination from conventional to contemporary image alignment techniques in remote sensing

Advancement in Remote Sensing Imaging sensors provides massive rise in the accessibility and quality of timely and repetitive multisensor, multimodal, multi band, multiview, multiresolution earth surveillance information on a global scale. Image alignment is a fundamental task in remote sensing. Though there is an extensive range of alignment techniques developed for different applications, the pipeline process involved in Conventional and Contemporary techniques is almost same, but the selection of algorithms involved in each stage of registration varies dynamically depending on the category of image, image acquisition sensors, Light source, viewpoint, time, their distortions, applications and domains. There is no universal method that holds good for the above said variations. This paper mainly aims to develop rule based blind alignment frameworks that can automatically map and select algorithm for the given pair of multisensory data. In the process of framework development a thorough comparative study of Multiple Conventional and Contemporary alignment technique is required. The key idea of the paper is to analyze the comparative study and identify and map the registration algorithm required for the given input image pair and a brief discussion of major alignment technique is given.