Infrared And Visible Image Fusion Based On PCNN And Intuitionistic Fuzzy Set

Nowadays,science and technology have developed rapidly,sensor technology has become more and more mature,and its applications in military and civilian applications have become more and more extensive.However,the information obtained by a single sensor is relatively single,low practicality and poor reliability.Under some relatively complicated conditions,it cannot meet the application requirements.By using the information fusion of multiple sensors,which can fuse the information obtained by different sensors,thereby it must obtain more reliable and comprehensive information and overcome the limitations and differences of a single sensor.The infrared and visible image fusion technology introduced in this paper is a classic example of multi-sensor fusion technology.It uses infrared and visible light sensors to obtain two images in the same target scene,and then uses the image fusion algorithm to fuse the acquired images.The fused image will get richer information,reduce the uncertainty and redundancy of information.The research background and significance of the subject,the current situation at domestic and abroad,and the three levels of image fusion are introduced first.Then,the imaging principles of infrared and visible images are described,as well as the two common fusion methods based on the spatial domain and the transform domain.Finally,the subjective and objective evaluation criteria are introduced for image fusion quality evaluation.Secondly,after a comparative analysis with wavelet transform and Contourlet transform,the non-downsampling contour wave transform(NSCT),which is more advantageous in the field of image fusion,is introduced.In view of the fact that theIHS color space is more in line with the observation characteristics of the human eye,this article uses the IHS transform to process the visible light image and extract its I component,and then combines it with the infrared image to perform image fusion by NSCT transform.After that,in the selection of the fusion algorithm of high and low frequency coefficients obtained by NSCT decomposition of visible and infrared images,this paper also focuses on the analysis.Because the low frequency sub-band contains the main energy information such as image contour and background,this paper uses the method of regional energy fetching for the low frequency part,and the PCNN can also effectively obtain the detailed information such as edges and texture.After simulation experiments in Matlab 7.0 environment and comparative analysis with other algorithms,it is found that the fusion effect is very ideal.Finally,in view of the uncertainty of information transmission during image fusion,this paper proposes the use of intuitionistic fuzzy set(IFS)theory for low frequency subband fusion.In view of the application of IFS theory,the choice of membership function and the conversion method of IFS to fuzzy set has a great influence on the fusion effect.This article uses various membership functions and IFS to fuzzy set conversion methods in the low-frequency fusion part.Compared with other algorithms,it is found that the image fusion effect is significantly improved when the algorithm in this paper adopts Gaussian membership function and difference correction method.