Block-Extraction and Haar Transform Based Linear Singularity Representation for Image Enhancement

Yingkun Hou, Xiaobo Qu, Guanghai Liu, Seong Whan Lee, Dinggang Shen

Research output: Contribution to journalArticle

Abstract

In this paper, we develop a novel linear singularity representation method using spatial K-neighbor block-extraction and Haar transform (BEH). Block-extraction provides a group of image blocks with similar (generally smooth) backgrounds but different image edge locations. An interblock Haar transform is then used to represent these differences, thus achieving a linear singularity representation. Next, we magnify the weak detailed coefficients of BEH to allow for image enhancement. Experimental results show that the proposed method achieves better image enhancement, compared to block-matching and 3D filtering (BM3D), nonsubsampled contourlet transform (NSCT), and guided image filtering.

Original languageEnglish
Article number6395147
JournalMathematical Problems in Engineering
Volume2019
DOIs
Publication statusPublished - 2019 Jan 1

Fingerprint

Image Enhancement
Image enhancement
Mathematical transformations
Singularity
Transform
Contourlet Transform
Image Filtering
Block Matching
Filtering
Experimental Results
Coefficient

ASJC Scopus subject areas

  • Mathematics(all)
  • Engineering(all)

Cite this

Block-Extraction and Haar Transform Based Linear Singularity Representation for Image Enhancement. / Hou, Yingkun; Qu, Xiaobo; Liu, Guanghai; Lee, Seong Whan; Shen, Dinggang.

In: Mathematical Problems in Engineering, Vol. 2019, 6395147, 01.01.2019.

Research output: Contribution to journalArticle

@article{f1d48810aaf640cfadaf0d20395ed3c8,
title = "Block-Extraction and Haar Transform Based Linear Singularity Representation for Image Enhancement",
abstract = "In this paper, we develop a novel linear singularity representation method using spatial K-neighbor block-extraction and Haar transform (BEH). Block-extraction provides a group of image blocks with similar (generally smooth) backgrounds but different image edge locations. An interblock Haar transform is then used to represent these differences, thus achieving a linear singularity representation. Next, we magnify the weak detailed coefficients of BEH to allow for image enhancement. Experimental results show that the proposed method achieves better image enhancement, compared to block-matching and 3D filtering (BM3D), nonsubsampled contourlet transform (NSCT), and guided image filtering.",
author = "Yingkun Hou and Xiaobo Qu and Guanghai Liu and Lee, {Seong Whan} and Dinggang Shen",
year = "2019",
month = "1",
day = "1",
doi = "10.1155/2019/6395147",
language = "English",
volume = "2019",
journal = "Mathematical Problems in Engineering",
issn = "1024-123X",
publisher = "Hindawi Publishing Corporation",

}

TY - JOUR

T1 - Block-Extraction and Haar Transform Based Linear Singularity Representation for Image Enhancement

AU - Hou, Yingkun

AU - Qu, Xiaobo

AU - Liu, Guanghai

AU - Lee, Seong Whan

AU - Shen, Dinggang

PY - 2019/1/1

Y1 - 2019/1/1

N2 - In this paper, we develop a novel linear singularity representation method using spatial K-neighbor block-extraction and Haar transform (BEH). Block-extraction provides a group of image blocks with similar (generally smooth) backgrounds but different image edge locations. An interblock Haar transform is then used to represent these differences, thus achieving a linear singularity representation. Next, we magnify the weak detailed coefficients of BEH to allow for image enhancement. Experimental results show that the proposed method achieves better image enhancement, compared to block-matching and 3D filtering (BM3D), nonsubsampled contourlet transform (NSCT), and guided image filtering.

AB - In this paper, we develop a novel linear singularity representation method using spatial K-neighbor block-extraction and Haar transform (BEH). Block-extraction provides a group of image blocks with similar (generally smooth) backgrounds but different image edge locations. An interblock Haar transform is then used to represent these differences, thus achieving a linear singularity representation. Next, we magnify the weak detailed coefficients of BEH to allow for image enhancement. Experimental results show that the proposed method achieves better image enhancement, compared to block-matching and 3D filtering (BM3D), nonsubsampled contourlet transform (NSCT), and guided image filtering.

UR - http://www.scopus.com/inward/record.url?scp=85071100516&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071100516&partnerID=8YFLogxK

U2 - 10.1155/2019/6395147

DO - 10.1155/2019/6395147

M3 - Article

VL - 2019

JO - Mathematical Problems in Engineering

JF - Mathematical Problems in Engineering

SN - 1024-123X

M1 - 6395147

ER -