A fractional differential fidelity-based PDE model for image denoising

In this paper, a new partial differential equation (PDE)-based model is proposed for image denoising. The new method is inspired by previous works in which the nonlinear diffusion approach obtained by using a coupling gradient fidelity term. Based on the long-term memory and nonlocality of fractiona...

Full description

Saved in:
Bibliographic Details
Published in:Machine vision and applications 2017-08, Vol.28 (5-6), p.635-647
Main Authors: Ma, Qianting, Dong, Fangfang, Kong, Dexing
Format: Article
Language:English
Subjects:
Accuracy
Communications Engineering
Computer Science
Image enhancement
Image Processing and Computer Vision
Mathematical models
Networks
Noise reduction
Original Paper
Partial differential equations
Pattern Recognition
Similarity
Vision systems
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, a new partial differential equation (PDE)-based model is proposed for image denoising. The new method is inspired by previous works in which the nonlinear diffusion approach obtained by using a coupling gradient fidelity term. Based on the long-term memory and nonlocality of fractional differential, we introduce a new fidelity term based on the combination of fractional-order fidelity term and global fidelity term to measure the similarity in the variation of images, which can prevent the staircase effect, and simultaneously enhance the noisy image, thus, the image becomes clearer and brighter. Numerical results are presented in the end to demonstrate that with respect to image denoising capability, our fractional fidelity-based model outperforms the gradient fidelity-based model.
ISSN:0932-8092
1432-1769
DOI:10.1007/s00138-017-0857-z