Image stitching method and device

Abstract

The invention provides an image stitching method, and the method comprises the steps: enabling to-be-processed images with the same focal length to be divided into one group, respectively carrying out noise reduction processing of each group of to-be-processed images, and obtaining the to-be-processed images after noise reduction; building a Gaussian pyramid for the to-be-processed images after noise reduction, building a DOG pyramid according to the Gaussian pyramid and a scale space, carrying out the positioning of feature points according to the DOG pyramid, and obtaining the feature points; carrying out the matching and stitching of each group of to-be-processed images after noise reduction according to the feature points, and obtaining a jointed image corresponding to each group of to-be-processed images; eliminating a joint gap of the image after stitching in a mode of weighted fusion, and obtaining to-be-fused images; extracting the feature points from all to-be-fused images, carrying out the matching of the feature points, obtaining the feature point pairs after the matching of the to-be-fused images, and carrying out the registration of the to-be-fused images according to the matched feature points; carrying out the fusion of the to-be-fused images after registration through employing a Laplace algorithm, and obtaining a fused image.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a method and device for image splicing. Background technique [0002] With the advancement of science and technology and the rapid development of digital imaging technology, digital imaging equipment has also been widely used. However, due to its own physical limitations, digital imaging devices cannot meet people's needs in some special application scenarios. For example, when images with a large field of view need to be acquired, the imaging device itself has a narrow field , the panoramic image cannot be obtained. The inventors found that when acquiring images with a large field of view, they usually rely on image stitching technology for processing. When only relying on image stitching technology to acquire images with large fields of view, the content of the acquired images is not clear, and some feature points are easily overlooked, thus As a result, ...

AI Technical Summary

Problems solved by technology

However, due to its own physical limitations, digital imaging devices cannot meet people's needs in some special application scenarios. For example, when images with a large field of view need to be acquired, the imaging device itself has a narrow field , the panoramic image cannot be obtained

The inventors found that when acquiring images with a large field of view, they usually rely on image stitching technology for processing. When only relying on image stitching technology to acquire images with large fields of view, the content of the acquired images is not clear, and some feature points are easily overlooked, thus As a result, the stitched image has low accuracy and cannot accurately reflect the features in the field of view

Images