SAR And Color Visible Images Fusion Algorithm Of Based On Region Division Strategy

Image fusion is an important branch of the study direction of image processing,which has the very high research value.For each sensor has its own limitations,image fusion organically combine the complementary information of two or more source image of the same scene in a certain algorithm to obtain a more accurate,rich and reliable image description.In order to eliminate the limitations of the monophyletic sensors,extracting the complementary information of the SAR image and color visible image to realize the fusion of SAR image and color optical image.The fusion results of the traditional fusion algorithm have the problems such as: spectral distortion,noise,structural texture fuzzy.The main work of this paper includes the following several aspects: study the research background and research status quo of image fusion;compare the imaging principle and image feature of SAR and optical images;summarize the traditional fusion methods of SAR and optical images,mainly study wavelet fusion algorithm;propose the fusion algorithm of SAR and color visible images based on region division strategy,based on the wavelet high frequency information of images we divide the space area into smooth region and feature region.In view of the differences between these two space regions,different processing strategies are used respectively for image fusion.For feature region,we effective fuse the targeted goals on the basis of keep the spectrum information.For smooth region,we directly maintain color visible image spectrum information.In order to verify the effectiveness of the proposed algorithm,we uses the spectral distortion(D),the structural similarity(SSIM),mutual information(MI),peak signal to noise ratio(PSNR),and other indicators to evaluate the fusion results.The experimental results show that the proposed algorithm not only contains clear structures,abundant spectral information but also inhabits the multiplicative speckle noise of SAR image in great degree.The fusion results are a great help to further image processing such as: image segmentation,feature extraction and target recognition,regional transformation testing etc.From the compare of the proposed algorithm and the combination of IHS and DWT fusion method,the proposed algorithm both in subjective and objective evaluation index has obvious advantages.