Destriping method using lifting wavelet transform of remote sensing image

Based on the characteristic of striping noise in remote sensing images, a new destriping noise technique for the improved multi-threshold method using lifting wavelet transform applied to remote sensing imagery is presented in this letter. Have used the lifting wavelet decomposition algorithm, the thresholds are determined by corresponding wavelet coefficients in every scale. Remote sensing imagery is so large that the algorithm must be fast and effective. The lifting wavelet transform is easily realized and inexpensive in computer time and storage space compared with the traditional wavelet transform. We also compare the method with some traditional destriping methods both by visual inspection and by appropriate indexes of quality of the denoised images. From the comparison we can see that the adaptive threshold method can preserve the spectral characteristic of the images while effectively remove striping noise and it did better than the existed ones.