Abstract:In this paper, we propose a novel algorithm to estimate Gaussian noise levels for captured natural images by rigorously analyzing the limiting distributions of the eigenvalue spectrum of a large covariance matrix with Gaussian samples. In order to select a relatively homogeneous region that best represents the noise, the corresponding image patches are first rearranged to construct a high-dimensional noise covariance matrix. And then, an optimal criterion for classifying homogeneous regions is derived based on the statistical relationship between the largest and the second largest eigenvalues of a sample covariance matrix. Moreover, we further explore the reasons for the bias of the maximum likelihood estimator of the noise variance both in high-dimensional settings and finite samples. According to random matrix theory, we clarify the asymptotic properties of the trace of a sample covariance matrix to measure the error bounds of estimation and then propose a new bias-corrected estimator. To this end, an effective estimation method for the noise level is devised based on the boundness and asymptotic behavior of pure noise eigenvalues of the selected patches. The estimation performance of our method has been guaranteed both theoretically and empirically. Experimental results have demonstrated that our approach can reliably infer true noise variance and is superior to the competing methods in terms of both estimation accuracy and robustness.

Efficient Noise Level Estimation Based on Principal Image Texture

No-Reference Image Noise Estimation Based on Noise Level Accumulation

Image noise level estimation based on affine reconstruction and noise sample histogram

Piecewise Noise Variance Estimation of Images Based on Wavelet Transform

A Single-Image Noise Estimation Algorithm Based on Pixel-Level Low-Rank Low-Texture Patch and Principal Component Analysis

Low-light Level Image De-Noising Algorithm Based on PCA

Single image noise level estimation by artificial noise.

Image Quality Assessment Based on Blurring and Noise Level

Noise Variance Estimation Method Based on Regression Analysis and Principal Component Analysis

Noise Estimation in A Noise-Adjusted Principal Component Transformation and Hyperspectral Image Restoration

Blind Additive Gaussian White Noise Level Estimation from a Single Image by Employing Chi-Square Distribution

Robust Estimation for Image Noise Based on Eigenvalue Distributions of Large Sample Covariance Matrices.

Efficient Image Sensor Noise Estimation Via Iterative Re-Weighted Least Squares.

Adaptive image noise level estimation with Chi-square distribution on the flat patches selected by improved PCANet and ResNet101

Segmentation Methods For Noise Level Estimation And Adaptive Denoising From A Single Image

Non-local image filter algorithm based on adaptive parameter regression

Image Denoising Using Quadtree-Based Nonlocal Means With Locally Adaptive Principal Component Analysis

Noise Estimation Using Statistics of Natural Images

Noise Estimation of Natural Images Via Statistical Analysis and Noise Injection

Robust Noise Estimation Based on Noise Injection