An Improved SAR Image Speckle Reduction Algorithm of Wavelet Threshold

This thesis proposes an improved SAR image speckle reduction algorithm of wavelet threshold. This method aims at overcoming the shortages of the discontinuity of the hard threshold function in the traditional wavelet threshold noise reduction method and the permanent bias between the estimated wavelet coefficient and composed wavelet coefficient in soft threshold function. Also on the basis of Garrote threshold function, it constructs a new self- adaption threshold function, which adds in 3 shape adjustment factors to make the function own the characteristic of shape visual adjustment. In MATLAB simulation experiment, X-band SAR river image shot by Scan SAR sensor of Radarsat-1satellite verifies the authentic capability of the improved algorithm to SAR image speckle reduction. Compared with the noise reduction effect of the traditional threshold function, the new threshold function is more effective in SAR image speckle noise reduction and has an excellent edge preservation capability.