Image splicing method and device and electronic equipment

Abstract

The embodiment of the invention provides an image splicing method and device and electronic equipment, and the method comprises the steps: determining a feature point matching pair between two to-be-spliced images; acquiring internal parameter matrixes and external parameter information of two cameras which shoot the two to-be-spliced images; calculating a first homography matrix between the two to-be-spliced images based on the acquired internal parameter matrixes and external parameter information; according to the first homography matrix, removing wrong feature point matching pairs from thefeature point matching pairs, and obtaining first feature point matching pairs; and splicing the two images to be spliced into a spliced image according to the first feature point matching pair and the first homography matrix. According to the invention, the first homography matrix is determined according to the property of the cameras, so that the wrong matching point pairs can be removed no matter the quantity is large or small, and the purpose of improving the splicing quality of the spliced image is achieved.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to an image splicing method, device and electronic equipment. Background technique [0002] At present, there is a need to splice two pictures to be spliced ​​into one spliced ​​picture. The technology of related mosaic images is: extract the feature points of the two images to be stitched respectively, and match the extracted feature points to obtain the matching pairs of feature points between the two images to be stitched. The wrong feature point matching pairs are removed from the point matching pairs to obtain the remaining feature point matching pairs, and the two images to be stitched are stitched into a stitching image based on the remaining feature point matching pairs. [0003] Due to the fact that when matching the extracted feature points, there will be false matching, and sometimes a large number of wrong feature point matching pairs will be generated....

AI Technical Summary

Problems solved by technology

[0004] Since the removal of wrong feature point matching pairs by the random sampling consistent method is only applicable to the case where there are few wrong feature point matching pairs, if the number of wrong feature point matching pairs is large, the random sampling consistent method may remove the correct Feature point matching pairs are removed, leaving wrong feature point matching pairs. At this time, the quality of image stitching will be seriously affected. Therefore, an image stitching method is urgently needed to improve the quality of image stitching.

Images