Denoising of Hyperspectral Remote Sensing Image using Multiple Linear Regression and Wavelet Shrinkage
Dong Xu, Lei Sun, Jianshu Luo · Proceedings of the 2013 International Conference on Information, Business and Education Technology (ICIBET 2013) · 2013
Hyperspectral remote sensing image is easily contaminated by noise, which will affect the application of hyperspectral image, such as target detection, classification and segmentation, etc.Therefore, a denoising method of hyperspectral remote sensing image based on multiple linear regression (MLR) and wavelet shrinkage (WS) is proposed.Firstly, the residual image and the predicted image are obtained via MLR.Secondly, WS is performed on the residual image to remove the noise in the spatial domain.Lastly, a final denoised image is obtained by the predicted image and the corrected residual image.The experimental results show that the proposed method can improve signal-tonoise ratio (SNR) of the hyperspectral image efficiently.