Research On Anti-Halo Method Based On Infrared And Visible Image Fusion

In recent years,the economic income and life quality of most people are gradually improving.In order to facilitate travel,many families buy private cars,so the number of cars running on the highway is increasing,and at the same time,it also brings more safety risks,one of the main reasons for safety problems is halo in the process of driving.In the process of meeting in the dark environment at night,the incorrect use of high beam will cause the phenomenon of excessive light brightness,which makes the opposite driver feel very dazzling,so it is difficult to observe the road condition information and deal with it quickly.This halo phenomenon will bring great trouble to the driver.To solve the above problems,the halo in the image can be eliminated through the fusion of infrared and visible image,and the image with higher definition can be obtained.This method is a better anti-halo way,which has practical significance in real life.This paper studies the anti-halo method based on the fusion of infrared and visible images,and the main contents are as follows:(1)The theory of infrared and visible image fusion is studied,and the process of infrared and visible image fusion is studied,including image preprocessing,image fusion and subjective and objective evaluation of image effect,and the correlation algorithm of image fusion is summarized and analyzed.(2)The anti-halo method based on LatLRR multi-layer decomposition is studied,aiming at the problems of incomplete halo elimination and insufficient detail information by the existing fusion methods of infrared and visible images,an anti-halo method based on Lat LRR combined with composite filtering multi-layer decomposition and significance detection is proposed.Firstly,HSV color space conversion is performed on visible images to extract the brightness component V of visible images.By Lat LRR combined with composite filtering,the brightness component V of visible images and infrared images are decomposed in multiple layers to obtain the corresponding low frequency layer and high frequency layer.The low frequency layer is fused with the average rule.A strategy based on significance detection and weight mapping was proposed to fuse the high frequency layer,and the two fusion layers were linearly superimposed to obtain V ’component,and then the H,S and V’ components were converted back to RGB space to obtain the resulting image with high resolution and halo elimination.By comparing with the experimental results of existing methods,the antihalo effect of the resulting image is analyzed by subjective and objective evaluation methods,and it is verified that the proposed algorithm has certain advantages in anti-halo field.(3)The image fusion method based on NSCT combined with guided filtering is studied,the infrared image and visible image are decomposed into high frequency layer and low frequency layer by NSCT algorithm.The low frequency layer is fused by weighted average rule based on guided filtering,and the high frequency layer is fused by large rule based on pixel value.Based on regional standard deviation and Gussian adaptive rules,the low level high frequency layer is fused.The feasibility of the algorithm is verified by using several groups of common images,and then the algorithm is used to fuse the images of the night halo scene,which verifies that the algorithm can be applied to the field of anti-halo,and the halo elimination effect is good.(4)The GUI interface of anti-halo system is designed.Based on the GUI interface function of MATLAB platform,the visual user interface of anti-halo system is designed based on the fusion of infrared and visible images.The system mainly includes image filtering,image enhancement and image fusion functions.The experimental results show that the algorithm has anti-halo effect,and the system integrates and implements the algorithm,which is more convenient for users to use.