The Remote Sensing Image De-noising Algorithm Based on the Wavelet Transform and 2D-PCA

Abstract

The wavelet transform methods are widely used in many remote sensing image de-noising fields, but there are still some problems, such as details missing and the large amount of calculation. Based on this, a new algorithm based on the wavelet transform and 2D-PCA is proposed. The algorithm firstly use wavelet decomposition to remote sensing image, then use 2D-PCA to deal with its high frequency coefficients to reduce the dimension, finally by using inverse wavelet transform to put on the low with high frequency coefficient. From the experimental results indicated that the new algorithm can get rid of the noise at the same time better keep relative data information, optimize the edge information and calculate less time.