The Research Of Remote Sensing Image Enhancement Algorithm Based On NSST

Remote sensing image plays an important role in the research and monitoring of the earth.Remote sensing has shifted from the military field to the civilian sector,thus making it more develop rapidly.However,in the process of acquisition and transmission of remote sensing image,it is affected by many factors,such as sensor,atmosphere,and so on.In order to facilitate the recognition of remote sensing image,the remote sensing image must be processed.Image enhancement is a kind of image processing method,which is mainly to highlight or strengthen some useful information in the image.Therefore,image enhancement is essential for the recognition of remote sensing images.In the paper,two new image enhancement methods are proposed to solve the problems of low contrast,low PSNR,weak edge preserving and missing details.One of them is a remote sensing image enhancement method based on adaptive threshold and guided filter in non-subsampled shearlet transform(NSST)domain.The guided filter is a image filtering algorithm,which has good smoothness and can keep the gradient of the image well.In view of these characteristics,guided filter is used to enhance the image detail and edge in the paper.Firstly,the original image is decomposed into a low-frequency component and several high-frequency components by NSST.Then a linear stretch is adopted for the low-frequency component to change the contrast of the image,while the adaptive threshold method is used to restrain the noise in the high-frequency components.After denoising,the guided filtering is used for dealing with them which improve the detail information and edge-gradient.The experimental results show that the proposed method can improve the visual effect of the remote sensing image,and compared with the other methods,it has a certain improvement in the information entropy,peak signal to noise ratio and structural similarity index in the objective indicators.Next,another remote sensing image enhancement method based on histogram equalization and guided filter in NSST domain is proposed.Histogram equalization is a classical algorithm for improving the contrast.It is used to preprocess for image toimprove the overall contrast of the image.a same linear transform method is dealt with low-frequency part after the decomposition of NSST;thresholding method is used for denoising the high-frequency parts,guided filter is same used to enhance the details and edge-gradient.Experimental results show that compared with the other methods,the proposed method can significantly improve the contrast of the image,enhance the image details and edge gradient.