Two-dimensional image registration method for partially matched images

Abstract

The invention discloses a two-dimensional image registration method for partially matched images. The method comprises the following steps: extracting the boundaries of two images for registration; conducting pre-registration on two boundary images through a geometrical characteristic difference distribution histogram to obtain rough spatial alternation parameters; correcting floating images with the rough spatial alternation parameters and removing redundant boundary points; performing precise registration with average Hausdorff distance to obtain accurate spatial alternation parameters, and repeating the steps till required precision is obtained. According to the two-dimensional image registration method for the partially matched images, the geometrical characteristic difference distribution histogram is combined with generalized Hausdorff distance, as a result, the influence of the redundant boundaries on a pre-registration result is reduced, the redundant boundary points in the floating images also can be removed, and accurate registration and correction of the images are realized.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a two-dimensional image registration method for partially matched images. Background technique [0002] Multiple images of the same object or similar objects reflect the characteristics of the object in different aspects, and a more comprehensive understanding of the object can be obtained by comparing, analyzing, and synthesizing these images. Due to different imaging equipment, time, imaging conditions, etc., the resulting images have spatial position changes (local or global). Before analysis and comparison, their spatial positions need to be aligned, that is, through appropriate spatial transformations, the images in the same The pixels at the same position reflect the same point in the object, which is the basic concept of image registration. The relatively immobile image I as the basis for registration is called M for the template image, but to be used with the I M Spatia...

AI Technical Summary

Problems solved by technology

But for the case where the boundaries do not completely correspond, how to get accurate registration results is a problem

[0005] The disadvantage of the method based on the average Hausdorff distance is that when the floating image has redundant boundaries compared to the template image, the redundant boundary points will have a greater impact on the similarity measure

Because the redundant part is removed, the remaining part of the floating image boundary can be precisely aligned with the corresponding boundary in the template image. At this time, in the average sense, the redundant part is relatively far away from the template image boundary point set, and the average distance value is large. If When the number of boundary points in the redundant part is relatively large, it will have a great impact on the objective function value of the optimization method, and the optimal registration result will not be obtained at the real registration position

Images