LAPB: Locally adaptive patch-based wavelet domain edge-preserving image denoising

•We present an edge preserving denoising technique based on wavelet transforms.•The multilevel decomposition of the noisy image is carried out.•Denoising performance is improved by aggregation. Image denoising is one of the most diversified research areas in the field of image processing and compute...

Full description

Saved in:
Bibliographic Details
Published in:Information sciences 2015-02, Vol.294, p.164-181
Main Authors: Jain, Paras, Tyagi, Vipin
Format: Article
Language:English
Subjects:
Benchmarking
Corners
Edge preservation
Image denoising
Local adaptive thresholding
Noise reduction
Preserves
Retarding
Wavelet
Wavelet transform
Wavelet transforms
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:•We present an edge preserving denoising technique based on wavelet transforms.•The multilevel decomposition of the noisy image is carried out.•Denoising performance is improved by aggregation. Image denoising is one of the most diversified research areas in the field of image processing and computer vision. It is highly desirable for a denoising technique to preserve important image features, such as edges, corners and other sharp structures of an image, after denoising. Wavelet transforms show excellent proficiency in providing efficient edge-preserving image denoising, due to their capability of suppressing noisy signals from an image. This paper presents a novel edge-preserving image denoising technique based on wavelet transforms. The multi-level decomposition of the noisy image is carried out to transform the data into the wavelet domain. A locally adaptive patch-based (LAPB) thresholding scheme is used to effectively reduce noise while preserving relevant features of the original image. Experimental results on benchmark test images demonstrate that the proposed method achieves competitive denoising performance in comparison to various state-of-the-art algorithms.
ISSN:0020-0255
1872-6291
DOI:10.1016/j.ins.2014.09.060