An image stitching method and device

Abstract

The embodiments of the present invention provide an image splicing method and device. According to the method, the overlapping area of a first image and a second image which are to be spliced is determined, and a plurality of pixel point pairs in the overlapping area are determined; a pixel point pair is optionally selected from the plurality of pixel point pairs as a first pixel point pair, and the first pixel point pair is projected onto a geometric curved surface of which the curved surface radius is a first preset value, and an arc distance L between the first pixel point pair is obtained;the curved surface radiuses of the remaining pixel point pairs are respectively calculated according to the depth values of target real scenes corresponding to the remaining pixel point pairs, a first depth value and the first preset value; the coordinate of each pixel point on an unfolded image is determined; and the coordinates of pixel points in the first image or the second image in the overlapping area are translated by L, so that two coordinates corresponding to each pixel point pair can be overlapped with each other; and the unfolded image in which the coordinates of the plurality of pixel point pairs are overlapped with each other is determined as a spliced image. With the image splicing method and device of the invention adopted, the coordinates corresponding to the pixel point pairs in the overlapping area can be completely overlapped with each other in the unfolded image, and the accurate splicing of the images to be spliced is realized.

Description

technical field [0001] The present invention relates to the technical field of image stitching, in particular to an image stitching method and device. Background technique [0002] Due to the limitations of shooting technology, it is generally difficult to use one image captured by one lens to reflect all the information of a large scene, so the image stitching technology that allows you to see more and wider images emerges as the times require. Image mosaic generally refers to a technique of stitching at least two images with overlapping pixel points into a large seamless image. Image stitching technology is widely used in security and virtual reality and other fields. [0003] The image stitching process mainly includes the following parts: stitching device structure design, geometric projection, image registration, image fusion and stitching. In the prior art, the stitching effect is mainly guaranteed by improving the image registration algorithm. [0004] Currently, th...

AI Technical Summary

Problems solved by technology

[0005] However, whether affine transformation or projective transformation matrix is ​​used for registration, the image registration algorithm describes the transformation relationship between the target pixels of the same depth of field, which shows that image registration is only applicable to the same or different depth of field of the target pixels. In small image stitching, for images with different depths of field of the target pixels, it cannot be stitched accurately

Images