Paper Info
Title
Fast and reliable noise level estimation based on local statistic.
Abstract
Flat patches are selected by using an iterative process.The preliminary estimated value is linearly correlated with the real noise level.It performs well for high frequency image because only real flat blocks are chosen.It is suitable for additive Gaussian noise.It can also process multiplicative Gaussian noise. Noise level is an important premise of many image processing applications. This letter presents an automatic noise estimation method based on local statistic for additive white Gaussian noise (WGN). Analysis of the distribution of local variance shows that when local variances are not greater than the threshold that satisfies a special condition, their average is always linearly correlated with the real noise variance. Thus the actual noise variance can be obtained from these patches. Based on this idea, this letter provides an iterative process to select flat blocks, and estimates noise variance from these homogeneous patches using principal components analysis. Addressing challenges in noise estimation has major contributions to (1) studies on the distribution of local statistic and (2) an iterative process for choosing flat patches, which is the fundamental work of patch-based methods. The experiment results show that the proposed algorithm works well over a large range of visual content and noise conditions, and performs well in multiplicative noise. Compared with several conventional noise estimators, it yields best performance and faster running speed.
Year
DOI
Venue
2016
10.1016/j.patrec.2016.03.026
Pattern Recognition Letters
Keywords
Field
DocType
White Gaussian noise,Additive noise,Multiplicative noise,Noise estimation,Principal components analysis
Value noise,Noise measurement,Pattern recognition,Salt-and-pepper noise,White noise,Artificial intelligence,Additive white Gaussian noise,Gaussian noise,Multiplicative noise,Mathematics,Gradient noise
Journal
Volume
Issue
ISSN
78
C
0167-8655
Citations 
PageRank 
References 
9
0.49
13
Authors
2
Name
Order
Citations
PageRank
Ping Jiang1172.34
Jian-Zhou Zhang2225.38