Open AccessWavelet denoising of multiframe optical coherence tomography data using similarity measures
Author(s) -
Habib Wajiha,
Sarwar Tabinda,
Siddiqui Adil Masood,
Touqir Imran
Publication year - 2017
Publication title -
iet image processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.401
H-Index - 45
eISSN - 1751-9667
pISSN - 1751-9659
DOI - 10.1049/iet-ipr.2016.0160
Subject(s) - wavelet , artificial intelligence , pattern recognition (psychology) , noise reduction , wavelet transform , speckle noise , computer science , similarity (geometry) , image quality , similarity measure , noise (video) , optical coherence tomography , non local means , speckle pattern , computer vision , mathematics , coherence (philosophical gambling strategy) , image (mathematics) , statistics , optics , image denoising , physics
Speckle noise is the main cause of image degradation in optical coherence tomography, which makes denoising an essential process to obtain quality images. This study proposes a wavelet‐based denoising technique in which detail coefficients are assigned weights using similarity measures of Pearson's correlation coefficient and structural similarity index (SSIM). Stationary wavelet transform is used for SSIM which is an image quality measure is used as optimisation criterion to denoise images in this study. Procedure of weight computation is discussed in detail. Average of these detailed components is used to denoise the images. Comparison of proposed technique with the existing techniques has been carried out at length. Extensive qualitative and quantitative analysis reveal that the proposed technique is efficient and performs better in terms of noise reduction while maintaining the structural contents of the image.