Image Matching Algorithm Based On ACT And Application Of Electronic Image Stabilization

Image matching is a fundamental problem in the areas of image process and machine vision.It has recently received a lot of attentions in a wide range of applications including pattern recognition,medical image analysis,remote sensing processing and electronic image stabilization.This thesis proposes a novel image matching method named affine consensus transform(ACT for short),which has borrowed the idea of VFC and focus on electronic image stabilization.The main works and contribution of this thesis are as follows.First,we introduce the basic principles of image matching and electronic image stabilization,and describe some classical image deformation technology and how image matching method works on electronic image stabilization.Then,a novel method named ACT,which is unlike the VFC algorithm and modeled by affine transform,is proposed for image matching.This method takes into account the image characteristics in electronic image stabilization and deals with the image matching and motion estimation problem through one step.Experiments show that this method is robust and can tolerate lots of fail matches.Finally,the thesis applies the proposed method to image stabilization.In this progress,the thesis analyzes the characteristics of the image in electronic image stabilization application and designs the framework of the electronic image stabilization system.In electronic image stabilization system,the serial images only have the transformation in vertical and horizontal,therefore,the proposed method should work well in the system,and the experiments verify the conclusion.