Infrared And Visible Image Fusion Based On Intuitionistic Fuzzy Sets In NSCT Domain

With the continuous innovation of emerging technologies,multi-source image processing technology is becoming more and more mature,which has attracted extensive attention and research in military,medical image,monitoring and other aspects.Generally,the information captured by a single sensor has been unable to meet the needs of our application to a certain extent,which is embodied in poor practicability and low reliability.Combine multiple sensors to complete the deficiencies and build a more comprehensive image system.Considering the differences between different images,this paper takes infrared and visible image fusion as a case study.For two images in the same environment,the fusion algorithm is used to fuse the images.After fusion,the fusion image will have the corresponding detailed features of infrared and visible images,making the fusion image information richer and reducing the redundancy of information.A color fusion image algorithm based on NSCT,HSV,PCNN and intuitionistic fuzzy set theory is proposed:1.Under the loss of edge and contour information,the fusion framework of multi-scale transformation is selected.The structure of LP transform,wavelet transform,contourlet transform and NSCT transform are analyzed,then the image is decomposed by the above transformation,and the decomposed images in different directions are analyzed,and the advantages and disadvantages of the fusion effect between them are compared,and the most advantageous NSCT transform is selected as the main fusion framework.2.A color image fusion algorithm with enhanced color image effect and HSV transformation is proposed.The purpose of this paper is to enhance the imaging effect of color image,through the analysis and comparison between IHS transform and HSV transform,and propose a rule to extract the V component of visible image by HSV transform.At the same time,the region energy maximum method is used to obtain the contour information and background information for the low frequency sub-band part,and the pulse coupled neural network(PCNN)is used to obtain the edge and texture information of the image.A large number of simulation experiments are carried out in MATLAB 7.0 environment.The results show that the fusion effect of the algorithm is remarkable,and the quality of the color image after fusion is effectively improved.3.A color image fusion algorithm based on HSV transform and intuitionistic fuzzy set theory is proposed.Aiming at the problem of fuzziness in the fusion process,this article introduces the intuitionistic fuzzy set theory(IFS)and applies it to the fusion rules of low frequency subband coefficients.This paper compares the influence of four methods of IFS to fuzzy set: mean method,proportion method,difference correction method and Gaussian membership function.Through a large number of simulation experiments,it is concluded that the color fusion image under Gaussian membership function is better;The fusion rule of PCNN transform is used for the high frequency subband coefficients,which effectively retains the details,contour information,background energy and other information of visible and infrared images.To sum up,subjective evaluation method and objective evaluation method are used to evaluate the fusion image,and compares with other color fusion algorithms in simulation experiments.It is concluded that the effect of image fusion based on NSCT domain combined with HSV transform,PCNN transform and improved intuitionistic fuzzy set theory is better and more significant,and in the MATLAB 7.0 environment,the data indicators have been significantly improved,so as to verify the effectiveness and feasibility of this algorithm.