A new approach for noise reduction in spine radiograph images using a non-linear contrast adjustment scheme based adaptive factor

Aouache Mustapha, Aini Hussain, Salina Abdul Samad

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

X-ray radiograph images are used usually for vertebral diseases detection and fractures that can be visible on lateral view. In general the x-ray images are poor quality images with low contrast and they do not provide momentous information concerning pathologies that are of interest to the medical researchers in terms of texture or colour. Consequently the enhancement may have a critical role in affording plenty and satisfactory visual information to the radiologist and clinician. In this paper, we propose a new approach for noise reduction in the cervical and lumbar radiograph images by employing a non-linear contrast adjustment scheme based adaptive factor. To achieve this main objective, firstly we investigate the use of non-linear gamma correction filter with variable gain factor value and draw an end conclusion of its effect to the quantum noise reduction in the spine radiographs images, secondly a new algorithm for adaptive gain factor detection was developed based on statistical pixel-level (SPL) features extraction and traditionally artificial neural networks (ANN's) model as a classifier to find "best" gain factor. Thirdly, experimental results are presented to examine and evaluate the filter performance, this evaluation done via visual interpretation and quantitative measurement by measuring the MSE and PSNR between the input and the resulting filtered images using gamma correction with adaptive gain factor versus different gain factor value.

Original languageEnglish
Pages (from-to)4246-4258
Number of pages13
JournalScientific Research and Essays
Volume6
Issue number20
Publication statusPublished - 19 Sep 2011

Fingerprint

spine
Noise abatement
noise reduction
Noise
X-radiation
Spine
adjusting
X-Rays
Quantum noise
disease detection
X rays
Neural Networks (Computer)
Pathology
neural networks
Image quality
Feature extraction
Classifiers
Color
Textures
researchers

Keywords

  • Feature extraction
  • Gamma correction
  • MLP model
  • Noise reduction
  • Spine X-ray images

ASJC Scopus subject areas

  • Agricultural and Biological Sciences(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Medicine(all)
  • Engineering(all)
  • Physics and Astronomy(all)

Cite this

@article{04b2d64437024319b7d83daab26ec7d0,
title = "A new approach for noise reduction in spine radiograph images using a non-linear contrast adjustment scheme based adaptive factor",
abstract = "X-ray radiograph images are used usually for vertebral diseases detection and fractures that can be visible on lateral view. In general the x-ray images are poor quality images with low contrast and they do not provide momentous information concerning pathologies that are of interest to the medical researchers in terms of texture or colour. Consequently the enhancement may have a critical role in affording plenty and satisfactory visual information to the radiologist and clinician. In this paper, we propose a new approach for noise reduction in the cervical and lumbar radiograph images by employing a non-linear contrast adjustment scheme based adaptive factor. To achieve this main objective, firstly we investigate the use of non-linear gamma correction filter with variable gain factor value and draw an end conclusion of its effect to the quantum noise reduction in the spine radiographs images, secondly a new algorithm for adaptive gain factor detection was developed based on statistical pixel-level (SPL) features extraction and traditionally artificial neural networks (ANN's) model as a classifier to find {"}best{"} gain factor. Thirdly, experimental results are presented to examine and evaluate the filter performance, this evaluation done via visual interpretation and quantitative measurement by measuring the MSE and PSNR between the input and the resulting filtered images using gamma correction with adaptive gain factor versus different gain factor value.",
keywords = "Feature extraction, Gamma correction, MLP model, Noise reduction, Spine X-ray images",
author = "Aouache Mustapha and Aini Hussain and {Abdul Samad}, Salina",
year = "2011",
month = "9",
day = "19",
language = "English",
volume = "6",
pages = "4246--4258",
journal = "Scientific Research and Essays",
issn = "1992-2248",
publisher = "Academic Journals",
number = "20",

}

TY - JOUR

T1 - A new approach for noise reduction in spine radiograph images using a non-linear contrast adjustment scheme based adaptive factor

AU - Mustapha, Aouache

AU - Hussain, Aini

AU - Abdul Samad, Salina

PY - 2011/9/19

Y1 - 2011/9/19

N2 - X-ray radiograph images are used usually for vertebral diseases detection and fractures that can be visible on lateral view. In general the x-ray images are poor quality images with low contrast and they do not provide momentous information concerning pathologies that are of interest to the medical researchers in terms of texture or colour. Consequently the enhancement may have a critical role in affording plenty and satisfactory visual information to the radiologist and clinician. In this paper, we propose a new approach for noise reduction in the cervical and lumbar radiograph images by employing a non-linear contrast adjustment scheme based adaptive factor. To achieve this main objective, firstly we investigate the use of non-linear gamma correction filter with variable gain factor value and draw an end conclusion of its effect to the quantum noise reduction in the spine radiographs images, secondly a new algorithm for adaptive gain factor detection was developed based on statistical pixel-level (SPL) features extraction and traditionally artificial neural networks (ANN's) model as a classifier to find "best" gain factor. Thirdly, experimental results are presented to examine and evaluate the filter performance, this evaluation done via visual interpretation and quantitative measurement by measuring the MSE and PSNR between the input and the resulting filtered images using gamma correction with adaptive gain factor versus different gain factor value.

AB - X-ray radiograph images are used usually for vertebral diseases detection and fractures that can be visible on lateral view. In general the x-ray images are poor quality images with low contrast and they do not provide momentous information concerning pathologies that are of interest to the medical researchers in terms of texture or colour. Consequently the enhancement may have a critical role in affording plenty and satisfactory visual information to the radiologist and clinician. In this paper, we propose a new approach for noise reduction in the cervical and lumbar radiograph images by employing a non-linear contrast adjustment scheme based adaptive factor. To achieve this main objective, firstly we investigate the use of non-linear gamma correction filter with variable gain factor value and draw an end conclusion of its effect to the quantum noise reduction in the spine radiographs images, secondly a new algorithm for adaptive gain factor detection was developed based on statistical pixel-level (SPL) features extraction and traditionally artificial neural networks (ANN's) model as a classifier to find "best" gain factor. Thirdly, experimental results are presented to examine and evaluate the filter performance, this evaluation done via visual interpretation and quantitative measurement by measuring the MSE and PSNR between the input and the resulting filtered images using gamma correction with adaptive gain factor versus different gain factor value.

KW - Feature extraction

KW - Gamma correction

KW - MLP model

KW - Noise reduction

KW - Spine X-ray images

UR - http://www.scopus.com/inward/record.url?scp=80053056184&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80053056184&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:80053056184

VL - 6

SP - 4246

EP - 4258

JO - Scientific Research and Essays

JF - Scientific Research and Essays

SN - 1992-2248

IS - 20

ER -