Image mosaic method and device and storage medium

Abstract

The invention provides an image mosaic method and device and a storage medium. The method comprises the following steps: positioning the overlapping region of two images to be stitched, obtaining theinformation of the overlapping region, dividing the two overlapping sub-images into blocks in the boundary line of the overlapping region, obtaining the pair of the sub-images of the overlapping region, and directionally registering the feature points to obtain the feature point set of the overlapping region; According to the optimized TPS thin plate spline function, the feature point set is calculated, two deformation images to be fused are obtained, the deformation sub-image pairs corresponding to the boundary line of the overlapping region in the deformation image to be fused are obtained,the pixel position information of the deformation sub-image pairs is mapped to the unit circle, and the smooth transition process is carried out, so as to obtain the overlapping region image. The invention extracts feature points in the overlapping area, which improves the efficiency of feature extraction and the accuracy of matching. Using the optimized TPS function to deform the original image,the problem of ghost and deformation is restrained effectively. Adaptively smoothing pixel value of unit circle solves the problem of chromatic gap.

Description

technical field [0001] The present invention mainly relates to the technical field of image processing, in particular to an image splicing method, device and storage medium. Background technique [0002] Image mosaic is an important branch of visual field and image processing. It seamlessly stitches more than two partially overlapping images to obtain images with wide viewing angles. [0003] The most stable of the existing image stitching methods is to use the SIFT (Scale Invariant Feature Transform) algorithm proposed by Lowe to extract and match image features. Although this algorithm has many advantages, its biggest drawback is that the real-time performance of the image stitching algorithm is poor, and a lot of useless consumption is wasted in the processing of images in non-overlapping areas, resulting in the existence of a large amount of useless information. Therefore, it is necessary to provide an image stitching method to overcome or alleviate the above technical ...

AI Technical Summary

Problems solved by technology

Although this algorithm has many advantages, its biggest disadvantage is that the real-time performance of the image stitching algorithm is poor, and a lot of useless consumption is wasted in the processing of images in non-overlapping areas, resulting in the existence of a large amount of useless information.

Images