The Research Of Remote Sensing Image Enhancement Algorithm Based On NSCT And NSST

With the development of science and technology.Remote sensing image is used as a special image to record geomorphologic features,and the application scope of remote sensing image is gradually expanded.Remote sensing images of imaging and the influence of some factors including atmospheric radiation transmission process,imaging equipment,etc.,these factors lead to the remote sensing image noise pollution,blurred vision,overall contrast decline,affect the quality of the remote sensing image,to increase the difficulty of remote sensing image processing.Image enhancement is an important means to targeted important details to improve image,weaken the unnecessary information contained in the image can be eliminated,and the noise,and improve the image resolution.Therefore,image enhancement plays an important role in the processing of remote sensing images.We propose two image enhancement methods for the problems of noise pollution,low resolution and weak edge details in remote sensing images.The First one,a remote sensing image enhancement method using a multidirectional morphological filter and adaptive unsharp masking in the NSCT domain.First,the initial image is decomposed by NSCT,and the low frequency subband and several high frequency subbands are obtained.Then,the coefficients of low-frequency subband are processed by linear transformation,and the overall contrast is improved.The high frequency subband is used to process the coefficients of the high frequency subband,remove noise pollution and enhance the contour information in all directions.Then,the inverse transformation of NSCT is used to reconstruct all subbands.Finally,the reconstructed image is used to improve the detail of the image using adaptive non-sharpening mask.The experimental results show that the visual effect of the algorithm is improved,and the objective indicators such as clarity,contrast,SNR and contrast algorithm are improved.The second method,a remote sensing image enhancement method based on NSST domain and guide the filter.Firstly,the original remote sensing image and the image obtained by adaptive histogram equalization have been decomposed by NSST,and their respective low frequency subbands and high frequency subbands are obtained respectively.Then,the low-frequency sub-band of the original image is processed by PM filtering to eliminate the noise of the low-frequency part and improve the contrast.For the high frequency part of the adaptive histogram equalization,the loop threshold method is used to make the noise,and then the high frequency subband of the denoising is filtered.In the end,all subbands are inversed by NSST.The experimental results show that the algorithm improves the visual effect of the image.Compared with the comparison algorithm,the signal-to-noise ratio and structural similarity are improved and the time is reduced.