Research On Intensity-based Image Registration Technology

Image registration is the process of overlaying two or more images taken from different devices or in different time.The goal is to find a best space transformation which can match the two images pixel in the same geometric position.Actually,it is a coordinates mapping process aiming to match the pixel of the sensed image to the reference image at the same position.Typically,registration is required in image recognition,image segmentation and image fusion and widely used in medical image processing,machine vision,target detection and so on.The main works in this paper are described as follows:Through summarizing the latest research achievements and development of image registration methods,we give the mathematical definition and the main steps of image registration.Research four area-based image registration methods that are usually used in the gray based image registration,and then analyses their performance,next implement an image registration algorithm based on both phase-correlation and cross-correlation.A novel image registration method based on mixed mutual information and improved particle swarm optimization algorithm was proposed,the improved particle swarm optimization algorithm based on Renyi's entropy was adopted firstly in global searching phase and using the Renyi entropy to computer the overlap mutual information in the two images.Then the mutual information measure based on Shannon's entropy was taken as the objective function after obtaining the global optimal solution vector while the Powell algorithm was used to obtain the optimal solution in local searching phase.The results showed that the proposed algorithm has the advantage in accuracy and effectiveness over other image registration methods.