An image fusion method, system and device

Abstract

The invention discloses an image fusion method, system and device, belonging to the technical field of image processing, comprising: performing smooth filtering processing on N source images respectively to obtain base layer images corresponding to the N source images; Make the difference with the corresponding base layer images respectively to obtain the corresponding detail layer images; build the weight maps corresponding to the N base layer images according to the N base layer images respectively; Perform pyramid decomposition and weighted average to obtain the fusion result of each base layer image after decomposition; reconstruct the fusion result of each base layer image after decomposition to obtain the fusion map of N source images. The present invention can reduce the complexity of the algorithm while ensuring that the image obtained by fusion does not appear halo phenomenon.

Description

technical field [0001] The present invention relates to the technical field of image processing, and in particular, to an image fusion method, system and device. Background technique [0002] Image fusion is a technology that fuses two or more images of the same scene from the same or different types of sensors to obtain a composite image. [0003] At present, pyramid is one of the most widely used image fusion methods. The process of using pyramid decomposition to fuse images is as follows: first, a weight map is established according to three weight factors such as contrast factor, saturation factor and exposure factor; Multi-frame source images and weight maps are decomposed into pyramids, and the decomposed images are weighted and averaged; finally, the fused pyramid images are reconstructed. [0004] However, the defects of the pyramid image fusion method are: First, when the pyramid image fusion method is used, it is generally necessary to decompose more than 5 layers...

AI Technical Summary

Problems solved by technology

[0004] However, the defects of the pyramidal image fusion method are as follows: First, when using the pyramidal image fusion method, the number of layers that needs to be decomposed generally needs to be more than 5 layers to achieve a good fusion effect, which makes the complexity and calculation amount of the algorithm relatively high. Big

The second is that since the weight map is constructed according to the weight factor, the relationship between light and dark in the source image will be reversed after the fusion process, and the halo phenomenon will appear in the obviously dark area on the edge of the bright area, which will seriously affect Fusion Image Effects

Images