Image Denoising Based on the One-Dimensional LPA-RICI Method

The recently proposed signal denoising method based on the relative intersection of confidence intervals (RICI) rule combined with local polynomial approximation is applied to image denoising, with images being transformed to one-dimensional signals. The obtained results, in the terms of peak signal-to-noise ratio (PSNR), are compared to the results obtained using the approach based on the intersection of confidence (ICI) rule. It is shown that the denoising approach using the LPA-RICI method outperforms the one using the LPA-ICI method. Although the proposed denoising approach does not outperform the state-of-the-art denoising techniques, the main advantages of the one-dimensional image denoising algorithm are its simplicity, and the fact that it is computationally less demanding than two-dimensional anisotropic image denoising approaches.