Single Image Noise Level Estimation Using Dark Channel Prior

Date

2019-09-22

Author

Yeşilyurt, Aziz Berkay
Erol, Aybüke
Kamışlı, Fatih
Alatan, Abdullah Aydın

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

143
views

0
downloads

Noise level is required as an input parameter in various image processing applications. In this work, we use the dark channel prior (DCP) to estimate the noise level of an image degraded by additive white Gaussian noise. We develop an approximate model of the probability density function of the dark channel of the noisy image. Using this model, the noise level is determined with the maximum likelihood estimation method from the dark channel intensity values of the noisy image. The results show that our method is faster than the state-of-the-art methods by about two orders of magnitude while providing slightly inferior estimation performance.

Subject Keywords

Noise estimation, Dark channel prior, Minimum of Gaussian variables, Maximum likelihood estimation

URI

https://hdl.handle.net/11511/40980

DOI

https://doi.org/10.1109/icip.2019.8803150

Collections

Department of Electrical and Electronics Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Blind Communications Sytem Structures For SIMO Channels Karakuetuek, Serkan; Tuncer, Temel Engin (2008-04-22) Frame based single input multi output (SIMO) communications systems are considered. In these systems frame length, offset, channel order and channel coefficients are all unknowns and there is no training session. Different system structures are considered for the estimation these unknowns blindly and best system structure is found. In addition a new method for channel order estimation is proposed.
Correlation tracking based on wavelet domain information Ipek, HL; Yilmaz, I; Yardimci, YC; Cetin, AE (2003-08-07) Tracking moving objects in video can be carried out by correlating a template containing object pixels with pixels of the current frame. This approach may produce erroneous results under noise. We determine a set of significant pixels on the object by analyzing the wavelet transform of the template and correlate only these pixels with the current frame to determine the next position of the object. These significant pixels are easily trackable features of the image and increase the performance of the tracker.
Rotation Calibration of Rigid Spherical Microphone Arrays for Multi-perspective 6DoF Audio Recordings Olgun, Orhun; Erdem, Ege; Hacıhabiboğlu, Hüseyin (2021-09-08) The preferred approach for multi-perspective six-degrees-of-freedom (6DoF) audio involves using multiple rigid spherical microphone arrays (RSMA) that can capture higher-order Ambisonics. RSMAs are spherically symmetric and allow the calculation of the local decomposition of sound fields over spherical harmonic functions. When multiple such arrays are used, multiple scattering occurs that can be equalized via methods that rely on multipole expansions if the positions of the arrays are known and the coordina...
Subarray selection in octave arrays Kaderoğlu, Ali Rıza; Çiloğlu, Tolga; Department of Electrical and Electronics Engineering (2022-2-04) Sensor layout of the octave array is designed to process a frequency range of several octaves. Array aperture of the highest octave band is limited to uniform line array (ULA) segment at the center. Forming a sparse array layout for the highest frequency band by using some of the remaining elements apart from ULA could bring some advantages such as higher array gain and better detection performance. Using a sparse array has a drawback of violating the spatial Nyquist limit, which may cause high sidelobes to...
Fine resolution frequency estimation from three DFT samples: Case of windowed data Candan, Çağatay (2015-09-01) An efficient and low complexity frequency estimation method based on the discrete Fourier transform (DFT) samples is described. The suggested method can operate with an arbitrary window function in the absence or presence of zero-padding. The frequency estimation performance of the suggested method is shown to follow the Cramer-Rao bound closely without any error floor due to estimator bias, even at exceptionally high signal-to-noise-ratio (SNR) values.

Citation Formats

A. B. Yeşilyurt, A. Erol, F. Kamışlı, and A. A. Alatan, “Single Image Noise Level Estimation Using Dark Channel Prior,” 2019, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/40980.