Method and system for image registration of diffeomorphic demons based on modality transformation

Abstract

The invention relates to a diffeomorphism demons image registration method and a diffeomorphism demons image registration system based on mode transformation. Firstly, affine transformation is performed on a reference image and a floating image, and then the mode transformation is performed on the reference image and the floating image; pixel gray value data in the images after the mode transformation is read, a deformation vector is obtained, space transformation is obtained through the deformation vector, and the deformation vector is updated to the space transformation; and after updating times reach first preset times and when mode transformation times reach second preset times, initial space transformation is used for the transformation of the floating image. According to the image registration method and the image registration system, advantages of two image registration methods of a mode transformation demons algorithm and a diffeomorphism demons algorithm in an image registration aspect are integrated, the initial space transformation passes through multiple iterations, so as to obtain relatively better initial space transformation, and when the relatively better initial space transformation is applied to the transformation of the floating image, image registration can be realized better. The method and the system are suitable for single-mode and multi-mode image registration and can be used for processing large or small deformation registration.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a method and system for image registration of differential homeomorphic demons based on modal conversion. Background technique [0002] Image registration is the process of matching and superimposing two or more images acquired at different times, different sensors (imaging devices) or under different conditions (weather, illumination, camera position and angle, etc.). It has been widely used. In remote sensing data analysis, computer vision, image processing and other fields. Simply put, image registration is an optimization process of spatial transformation or mapping that maximizes the similarity between two images. [0003] Image registration can be divided into two categories: single-modal registration and multi-modal registration. The images in single-modal registration are from the same imaging device, while the images in multi-modal registration are from different imaging...

AI Technical Summary

Problems solved by technology

[0007] Based on this, it is necessary to address the problems of single function and insufficient processing capacity of the existing image registration methods, and provide a modality conversion-based Image registration method for diffeomorphism demons

Images