A sparse representation-based image fusion method guided by high-frequency information

Abstract

The invention relates to an image fusion method based on sparse expression and guided through utilizing the high frequency information. The method comprises steps that step 1, the image information isdecomposed into a high frequency portion and a low frequency portion, a non-downsampling pyramid decomposition mode is employed to extract the high frequency information of the image information in different levels; step 2, the high frequency information and the low frequency information of the sample image are utilized to respectively accomplish establishment of an over-complete dictionary suitable for the high frequency information and establishment of an over-complete dictionary suitable for the low frequency information, and the over-complete dictionary is acquired through an instance learning method; step 3, the high frequency information is utilized to guide fusion of the low frequency information; step 4, an absolute maximum value is employed as the fusion strategy to carry out fusion of the high frequency information to acquire the high frequency information in each level after fusion; and step 5, fusion image reconstruction is carried out. The method is advantaged in that aproblem caused by insensitivity to the high frequency information after image sparse expression is ameliorated, the block effect generated during fusion is inhibited, and the relatively high quality fusion result can be acquired for various types of image fusion.

Description

【Technical field】 [0001] The invention relates to an image fusion method based on sparse representation guided by high-frequency information. Image fusion has wide application in the field of image application and belongs to the technical field of digital image processing. 【Background technique】 [0002] Image fusion refers to the technology of integrating multi-source image information or image information under different conditions of the same source into a fused image. The essence of fusion is to enable image information to overcome the defects of single sensor imaging, so that the fused image can meet the requirements of human eye recognition and Computer follow-up processing requirements. Image fusion can make the fused image extract more effective information from various source images, which facilitates more accurate analysis and application of images. In order to fully mine and utilize all kinds of complex data, image fusion technology has developed rapidly. Image ...

AI Technical Summary

Problems solved by technology

[0005] 1. Purpose: The purpose of the present invention is to provide a sparse representation-based image fusion method guided by high-frequency information, to solve the weak performance of edges and the limitation of the maximum 1 range in traditional fusion methods based on sparse representation. The mismatch of high and low frequency information caused by the number fusion strategy leads to a block effect in the fusion result, which leads to problems such as fusion image distortion and image quality degradation.

Images