Research On Image Fusion Algorithm Of Infrared And Visible Image

lmage fusion technique is an image proeessing techniqe whieh put the image data ofthe same seene in different spectrum or focuses into one image.The fused image can make abetter visibility and contain complementary information of source images.This teehnique can integrate the advantage of multi Ple imaging deviees.Visible and infrared image fusion teehnique was often used,for visible image is rich in detail,at the same time,infrared image can provide clear outline of objeet and also ean find hidden object.That means they have com plementary information.visible and infrared image fusion teehnique has a wide range of applications,and also has great value at battlefield auxiliary at night and ear navigation.This dissertation is aimed at fusing the visual image and infrared image about asame scene into a single image to display both images1 information.According towavelet transform(Mallat algorithm),the visual image and infrared image are bothtransformed into several child-band images which carry different information among.Infrared image and visible image reflect on different shooting instrument,different characteristics of target scene.Firstly,wavelet decomposition to gets high frequency coefficient and low-frequency coefficients,high frequency and low frequency respectively.The high frequency coefficients use the rule of neighborhood regional correlation variance,the low frequency coefficients adopt the rule of gradient weighted sum.Then,the fusion image is gotten by inverse transform.Finally,image fusion effect is comprehensively evaluated by information entropy,average gradient and mutual information.Experimental results show that the fusion image contains more information,which is more advantageous to human eye judgment and observation.