Research On The Theory Of Image Registration And Registration Algorithm

0Registration is a fundamental task in image Processing used to (geometrically) match two or more images, for example, at different time, from different sensors or from different viewpoints. Registration is the foundation of many image processing and applications such as object recognition, change detection, 3-D modeling, and the effect of registration directly works on the following processing steps.There are two types in image distortion: those which are the source of misregistration and those which are not the source of misregistration, image registration is to find proper technique which is suitable for the former while eliminate the effect of the latter. There are a variety of registration methods, but none of them are adapted to all images, so how to obtain a proper method to the image we are working on is the main task on image registration.This paper aims at pilot study on the concept and signification of registration, and common methods and models of registration. Moreover, it also reviews several conventional registration methods. The main task of this paper is the research of three image registration method.Firstly, by mading some analyses about image registration method based on control point, achieved two images registration. The experiments show that in ERDAS IAMAGINE software system, image registration method based on control point is effective in image registration, and it has higher registration precision.Secondly, discusses an extension of the well-known phase correlation of FT technique to cover translation, rotation and scaling in image registration. At the same time, it presents a new improved algorithm, which is characterized by representing the rotation and scale as a translational movement with polar coordinates, and it improves the computational speed of the phase correlationmethod. The experiments show that this algorithm is effective in image registration, and it has higher registration precision and computation efficiency.Thirdly, optimization algorithm is studied: we presented a new algorithm which combines the advantages of both local optimization algorithm (simplex search method) and whole optimization algorithm (simulated annealing algorithm), which not only overcome the problem of local extremum, but also improve the speed and precision of registration. The innovation of the research is that it ameliorated the simulated annealing algorithm, and combine local optimization algorithm (simplex search method), presented a new efficient whole optimization algorithm. The experimental results indicate the exact and rapid, robust merits of image registration by using above methods, which can meet well practical applied demands.Fourthly, after image registration, the evaluations of registration algorithm were given. We employed the technique of mosaic to pick them up, and formed the panoramic picture.Finally, the author concluded the work and give out some suggestions for further research.