relation: https://khub.utp.edu.my/scholars/943/
title: Image SNR estimation using the autoregressive modeling
creator: Kamel, N.
creator: Kafa, S.
description: A number of techniques have been proposed during the last two decades for Signal-to-Noise Ratio (SNR) estimation in images. The majority of these techniques are based on the cross-correlation function of two images of the same area. However, the need for two images to estimate SNR value confines these techniques to non-stored images and thus limits their applications. In this paper the second order statistics of image corrupted by additive white noise are modeled by Autoregressive-model and the relationship between AR model and linear prediction is utilized in estimating the predictor coefficients. The predictor is then used to estimate the zero-offset autocorrelation value and accordingly obtain the power of the noise-free image. Unlike others, the proposed technique is based on single image and offers the required accuracy and robustness in estimating the SNR values.
date: 2010
type: Conference or Workshop Item
type: PeerReviewed
identifier:   Kamel, N. and Kafa, S.  (2010) Image SNR estimation using the autoregressive modeling.  In: UNSPECIFIED.     
relation: https://www.scopus.com/inward/record.uri?eid=2-s2.0-79952753385&doi=10.1109%2fICIAS.2010.5716130&partnerID=40&md5=a90515c1da48b3f0a3bc0a2eafbbf121
relation: 10.1109/ICIAS.2010.5716130
identifier: 10.1109/ICIAS.2010.5716130