Speckle reduction method for thyroid ultrasound images in neutrosophic domain
Open Access
- 1 February 2016
- journal article
- research article
- Published by Institution of Engineering and Technology (IET) in IET Image Processing
- Vol. 10 (2) , 167-175
Abstract
Neutrosophy is a useful tool for handling uncertainty associated with the images and widely used in image denoising. Speckle noise is inherent in ultrasound images, which generally tends to reduce resolution and contrast, thereby degrading the diagnostic accuracy. This paper presents a variational method based on Gamma distribution in the neutrosophic domain to improve clinical diagnosis and to enhance quality of ultrasound images. In this method, image is transformed into the neutrosophic (NS) domain via three membership subsets ( true , indeterminate and false ). Then, the filtering operation is applied based on total variation regularisation to reduce the indeterminacy of the image, which is measured by the entropy of an indeterminate set. The proposed speckle reduction method has been assessed on both the artificial speckle simulated images and real US images. The experimental results reveal the superiority of the proposed method in terms of both quantitatively and qualitatively as compared to other speckle reduction methods reported in the literature. Furthermore, the visual evaluation of despeckled images demonstrates that the proposed method suppresses the speckle noise as well as preserves the textures and fine details.Keywords
This publication has 23 references indexed in Scilit:
- Efficient Nonsmooth Nonconvex Optimization for Image Restoration and SegmentationJournal of Scientific Computing, 2014
- Nonconvex sparse regularizer based speckle noise removalPattern Recognition, 2013
- Speckle reduction with edges preservation for ultrasound images: using function spaces approachIET Image Processing, 2012
- Removing Multiplicative Noise by Douglas-Rachford Splitting MethodsJournal of Mathematical Imaging and Vision, 2009
- Adaptive total variation image deblurring: A majorization–minimization approachSignal Processing, 2009
- A New Total Variation Method for Multiplicative Noise RemovalSIAM Journal on Imaging Sciences, 2009
- A Variational Approach to Removing Multiplicative NoiseSIAM Journal on Applied Mathematics, 2008
- Multiplicative Denoising and Deblurring: Theory and Algorithms2006
- A novel multiscale nonlinear thresholding method for ultrasonic speckle suppressingIEEE Transactions on Medical Imaging, 1999
- Adaptive speckle reduction filter for log-compressed B-scan imagesIEEE Transactions on Medical Imaging, 1996