Research On Infrared And Visible Image Fusion Based On NSST

With the rapid development of image sensor technology,more and more methods and tools can be used to acquire multi-sensor image data.However,due to the complexity and diversity of data acquired by sensors,how to make full and effective use of these data becomes an urgent problem.Among them,infrared and visible image fusion is a key research field.Visible light sensor has low cost and high resolution,but it is affected by external environmental factors.Infrared sensors cut through darkness and smoke,and are less affected by external light,capturing thermal targets.Combining information from the two sensors can make the image richer and contain more valuable features.In recent years,the image fusion method based on multi-scale transformation has attracted more and more scholars’ attention,because it can highlight the important details and features of the image,and has become a hot research direction in the field of image fusion.In this paper,the fusion of infrared and visible images based on non-subsampled shear wave transform(NSST)is studied.In view of the traditional fusion algorithm can not retain image details well and the poor adaptability of the fusion method,the corresponding improved fusion method is proposed.The main contents of this paper are as follows:1.In this paper,the realization principle of Laplace pyramid transform,wavelet transform,double-tree complex wavelet transform,NSCT transform and NSST transform is deeply studied,and the infrared and visible image fusion framework based on multi-scale transform method is introduced in detail.Then,in order to compare the decomposition ability of various multi-scale transform methods,this paper uses different multi-scale transform methods to decompose the same level of the same image,analyzes the differences in the decomposition direction and performance of different multi-scale transform methods,and finally verifies the multi-direction and flexibility of NSST transform decomposition image,which lays a foundation for the research of the following chapters.2.A fusion method of infrared and visible images based on NSST and phase consistency was proposed in order to better obtain the texture and contour of scene objects.Firstly,NSST transform is used to decompose infrared and visible images.For low-frequency sub-band images of images,features in the images are effectively extracted by PC algorithm,and then the final low-frequency coefficient is obtained by weighted fusion combined with energy significance graph.For the high frequency subband image,the high frequency coefficient is obtained by improving the traditional Laplacian operator and combining the local energy.Finally,the fused image is obtained by NSST inverse transformation.3.Compared with CVT,MSVD,ADF and FPDE,the experimental comparison on MSRS data set and TNO data set respectively proves that the proposed algorithm can retain more source image information into the fused image,and make the target area more significant and the image fusion quality better,which verifies the effectiveness and robustness of the proposed algorithm.