Poissonian Hyperspectral Image Denoising without Using Anscombe Transform

In the most existing Hyperspectral Image (HSI) denoising methods, Poisson noise is first transformed into Gaussian noise through Anscombe transform and then remove it. However, transform errors may occur that affect the final denoising results when using Anscombe transform. In this paper, we propose a Poissonian hyperspectral image denoising method without using Anscombe transform (WUAT) to directly remove the noise of Poissonian HSI under the maximum a posteriori (MAP) model by finding the minimum value of the negative logarithmic Poisson log-likelihood combined with the total variation (TV). The experimental results show that the proposed method can acquire better performance than most state-of-the-art denoising methods.