Image registration method based on similar feature triangles

Abstract

The invention relates to the field of digital image processing, and provides an image registration method to provide an accurate and quick solution for the image registration under similar transformation. The adopted technical scheme is that the image registration method based on similar feature triangles comprises the following steps: 1) extracting and matching feature points; 2) randomly selecting samples to form feature triangles; 3) pre-checking the similar feature triangles; 4) selecting sample points according to a maximum Euclidean distance method; 5) obtaining a transformational model through transformational model solving and all-data checking. The method is mainly applied to the digital image processing.

Description

technical field [0001] The invention relates to the field of digital image processing, in particular to an image registration method under similar transformation. technical background [0002] As the core technology of image stitching, image registration has important applications in many aspects such as panoramic stitching, super-resolution images, motion estimation of two-dimensional planar objects, and distributed camera network calibration. The purpose of image registration is to find a transformation model that can describe the relationship between two images according to the overlapping regions of images collected from different perspectives. The accuracy and speed of image registration will have a direct impact on the subsequent image processing. When imaging a two-dimensional scene, the transformation relationship sought by image registration can be divided into rigid transformation, similar transformation, affine transformation and projective transformation. Among...

AI Technical Summary

Problems solved by technology

[0005] However, the image registration method based on traditional RANSAC cannot meet higher requirements in the estimation of the transformation model, accuracy and speed, especially in the image registration of similar transformations with wide applications.

Using the standard RANSAC algorithm for complex projection transformations will waste unnecessary time and resources in the estimation of the similarity transformation model. At the same time, the full data inspection will cause a lot of unnecessary waste in the iterative process of the algorithm due to the discarding of unsuccessful samples. time, and the randomness of the algorithm may cause the Euclidean distance between some selected data points to be too small, making the estimated transformation model unstable and inaccurate

Images