The Research And Implementation Of Image Registration System

Image registration, which is the first issue should be settled in image data amalgamation and applied in the aspects of remote sensing image processing,medical image analysis and so on, is all significant task in the fields of computer vision and mode identification.There will be relative move,rotation and different proportion zoom between the same image data gained at different time using different sensor and imaging mode,So we must settle the image registration first of all.This paper did research on image registration in the same scene. The main works are detailed as followed:1、Study the concepts, principles and common methods of gray-based registration. Implemented gray-based image registration based on Powell optimization algorithm and PSO algorithm.2、We study several extraction operators of feature points. Analyze performance of Harris operator, SUSAN operator, SIFT operator, by comparing the detection efficiency, rotation invariance and the performance of anti-noise.And we give some conclusions according to the analysis. Improve the conventional Harris algorithm and SIFT algorithm so that the accuracy and efficiency of feature point matching is improved as well.3、According to the analysis above, we present the manual image mosaic and the automatic image mosaic. The advantages of manual registration are the less demanding on the accuracy of regis- tration points and its flexibility; the automatic registration needs no human intervention.4、Image registration system is present. For different images, we can choose different registration strategies, which make this system more flexible and extensible.