Multi-focus image fusion method with good anti-noise environment effect

Abstract

The invention discloses a multi-focus image fusion method with a good anti-noise environment effect. The method comprises the following fusion methods of S1, obtaining focus information of source images through focus detection; S2, obtaining focus areas corresponding to the source images through utilization of an image matting technology according to the focus information, improving contact between the focus areas and adjacent pixels of the images, and effectively solving the fusion problem of multi-focus images comprising different content in the same scene; and S3, fusing all focus areas toform a fused image. According to the method, detailed detection is carried out on the focus images fully; noises are eliminated; the contact between the focus areas and the adjacent pixels of the images is fully improved; through utilization of a high precise detection algorithm, noise impurities in a fusion process are eliminated; and moreover, the method has the advantages that time and space resources required by computing are few and the image fusion quality is high.

Description

technical field [0001] The invention relates to the technical field of image fusion processing, in particular to a multi-focus image fusion method with good anti-noise environment effect. Background technique [0002] A single sensor of a camera cannot focus on all objects. Therefore, when imaging different objects in the same scene, the camera cannot obtain clear images of all objects in the scene. Multi-focus image fusion technology can fuse multiple images into one image and provide more complete, accurate and reliable information for human observation or subsequent processing by machines. Multi-image fusion algorithms have gained more and more applications in remote sensing image processing, machine vision and other fields. [0003] At present, the research on multi-focus image fusion algorithms is mainly divided into two categories: transform domain fusion and space domain fusion. Most of the existing transform domain fusion algorithms are based on multi-scale transfo...

AI Technical Summary

Problems solved by technology

[0008] In step S2, according to the focus information, the focus area corresponding to the source image is obtained by matting technology, which enhances the connection between the focus area of ​​the image and adjacent pixels, and effectively solves the multi-focus image fusion problem with different contents in the same scene;

Images