Panoramic image mosaic method, intelligent terminal and storage medium

Abstract

A method for mosaic panoramic images, an intelligent Terminal and a storage medium are disclosed. The method comprises: since the user takes panoramic photographs using a relatively fixed panoramic photographing apparatus, in panoramic shooting, the rotation angle of the camera can be easily obtained or at least roughly measured, and the parameters which can not be directly measured, such as the internal parameters of the camera, can be roughly calibrated on line to obtain the predicted value. Combined with the rotation matrix of the rotation angle of the camera, a global homography matrix which is estimated in advance is constructed. In panoramic image mosaic, the parameters in homography matrix are adjusted by feature point matching pair and feature line matching, and the problem of unsuccessful mosaic caused by the shortage of feature point matching pair is remedied by using line matching pair. The homography matrix of the photographs which occasionally appear without enough featurepoints and feature lines is calculated by using the homography matrix of the photographs adjacent to the photographs to ensure the success of the stitching.

Description

technical field [0001] The present invention relates to the technical field of image stitching, in particular to a panoramic image stitching method based on an estimated homography matrix, an intelligent terminal and a storage medium. Background technique [0002] Image mosaic technology is the technology of stitching several images with overlapping parts into a large seamless high-resolution image. The so-called panorama in panoramic image mosaic means that the image has a relatively wide field of view. A single image is fused into a single image. Specifically, panoramic stitching is to use multiple cameras to shoot the same scene at different angles of view (you can also use one camera to shoot at different times and different angles of view), and then correct, denoise, match and fuse the obtained multiple images. , and finally construct an image with high definition, smooth edges and high resolution. [0003] With the development of computer vision technology and its po...

AI Technical Summary

Problems solved by technology

The current mainstream panorama image stitching algorithms are almost all based on image feature points. Without enough feature points, effective stitching cannot be performed, resulting in a low average stitching success rate; and in some cases, stitching based on feature points results cannot be guaranteed

Images