Research On Image Fusion Algorithm Base On Non-subsampled Contourlet Transform

Image fusion is an information fusion technology widely used in image understanding and computer vision.Image fusion technology can effectively reconstruct image feature information and combine them into high-quality images.It has been widely applied in military,remote sensing,medical science and other fields have a wider range of applications.Based on the research of image enhancement related algorithm and image fusion algorithm theory and technology,this paper focuses on the analysis of the fuzzy logic algorithm for visible light image enhancement and the guide filtering algorithm for infrared image enhancement.On the Basis,we propose an infrared image enhancement based on the Canny edge arithmetic to improve the guided filtering,and it is with the nonsubsampled contours transform.The combination is used for image fusion.The main work of this paper is as follows:1.Improve the guided filtering algorithm.It is used to solve the problem of over smoothness caused by traditional guided filtering algorithm in enhancing infrared images.Firstly,the edge detection of infrared image is performed by Canny edge detection operator.Then,the local gradient is replaced by the absolute gradient amplitude of the local Canny edge detection operator after non-maximum suppression and double-threshold processing,and the edge weighting factor of the guide filter is redefined.Experiments show that this method can effectively solve the image smoothing phenomenon and improve the image fusion effect.2.An image fusion method combining non-subsampled contourlet transform and image enhancement is proposed to solve the problems of serious loss of infrared detail information,blurring of edges,and insufficient contrast of visible light images in the fusion of visible light and infrared images.First,the fuzzy logic algorithm is used to enhance the visible light image to enhance the contrast and highlight the effective information of the image.Then the Non-subsampled contourlet transform decomposition of the enhanced visible light and infrared image is performed to obtain the low frequency and high frequency subbands;then the high frequency subbands of the infrared image are adopted.The improved guide filter enhances the details such as edges;finally,the fused image is obtained by inverse transform of NSCT.The experimental results on different datasets show that this method solves the problem of insufficient information on edges and detailed information after fusion of visible light images and infrared images,and the fusion images need to be evaluated on information entropy,standard deviation,and mutual information and other evaluation indicators.Better than other methods.