Application Of Second-Order Histograms In Noise Reduction And Contrast Measurement Of Intensity Images

This research focuses on a type of information provided by the pixels in intensity images. The types of images that are the focus of this research are scanning electron microscopy (SEM) and magnetic resonance imaging (MRI) images. These two types of images were chosen for the amount of details that...

Full description

Saved in:

Bibliographic Details
Main Author:	Chan, Wai Ti
Format:	Thesis
Published:	2019
Subjects:	TA1501-1820 Applied optics Photonics
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my-mmu-ep.7767
record_format	uketd_dc
spelling	my-mmu-ep.77672020-09-22T17:53:39Z Application Of Second-Order Histograms In Noise Reduction And Contrast Measurement Of Intensity Images 2019-10 Chan, Wai Ti TA1501-1820 Applied optics. Photonics This research focuses on a type of information provided by the pixels in intensity images. The types of images that are the focus of this research are scanning electron microscopy (SEM) and magnetic resonance imaging (MRI) images. These two types of images were chosen for the amount of details that they can present using intensity values. In turn, the details that they provide lead to the use of the secondorder derivatives of the values of pixels. The second-order derivatives provide further information about the images, specifically their levels of noise and contrast. The second-order derivatives are obtained from applying a Laplacian operator on the pixels in the image. The second-order derivatives of the pixels can be collected into histograms, which are henceforth referred to as “second-order histograms”. These histograms have shape profiles with certain characteristics that are common to all images with adequate contrast and low levels of noise, e.g. profiles that approximately resemble Laplace distribution profiles. The main theory in this research is that any deviation from these presence is an indicator of the presence of factors that affect the quality of an image. One of the characteristics of the Laplace distribution profile is the convergence of two concave and non-contiguous curves into a sharp peak. The peak contains data on image details that have uniform intensity, such as regions of the same material in SEM images. Therefore, the absence of the peak in the second-order histogram of an image indicates the presence of factors that affect those details, such as noise. This is the idea behind a method to reduce noise through the restoration of the Laplace distribution profile by changing the pixel values of an image. 2019-10 Thesis http://shdl.mmu.edu.my/7767/ http://library.mmu.edu.my/library2/diglib/mmuetd/ phd doctoral Multimedia University Faculty of Engineering & Technology
institution	Multimedia University
collection	MMU Institutional Repository
topic	TA1501-1820 Applied optics Photonics
spellingShingle	TA1501-1820 Applied optics Photonics Chan, Wai Ti Application Of Second-Order Histograms In Noise Reduction And Contrast Measurement Of Intensity Images
description	This research focuses on a type of information provided by the pixels in intensity images. The types of images that are the focus of this research are scanning electron microscopy (SEM) and magnetic resonance imaging (MRI) images. These two types of images were chosen for the amount of details that they can present using intensity values. In turn, the details that they provide lead to the use of the secondorder derivatives of the values of pixels. The second-order derivatives provide further information about the images, specifically their levels of noise and contrast. The second-order derivatives are obtained from applying a Laplacian operator on the pixels in the image. The second-order derivatives of the pixels can be collected into histograms, which are henceforth referred to as “second-order histograms”. These histograms have shape profiles with certain characteristics that are common to all images with adequate contrast and low levels of noise, e.g. profiles that approximately resemble Laplace distribution profiles. The main theory in this research is that any deviation from these presence is an indicator of the presence of factors that affect the quality of an image. One of the characteristics of the Laplace distribution profile is the convergence of two concave and non-contiguous curves into a sharp peak. The peak contains data on image details that have uniform intensity, such as regions of the same material in SEM images. Therefore, the absence of the peak in the second-order histogram of an image indicates the presence of factors that affect those details, such as noise. This is the idea behind a method to reduce noise through the restoration of the Laplace distribution profile by changing the pixel values of an image.
format	Thesis
qualification_name	Doctor of Philosophy (PhD.)
qualification_level	Doctorate
author	Chan, Wai Ti
author_facet	Chan, Wai Ti
author_sort	Chan, Wai Ti
title	Application Of Second-Order Histograms In Noise Reduction And Contrast Measurement Of Intensity Images
title_short	Application Of Second-Order Histograms In Noise Reduction And Contrast Measurement Of Intensity Images
title_full	Application Of Second-Order Histograms In Noise Reduction And Contrast Measurement Of Intensity Images
title_fullStr	Application Of Second-Order Histograms In Noise Reduction And Contrast Measurement Of Intensity Images
title_full_unstemmed	Application Of Second-Order Histograms In Noise Reduction And Contrast Measurement Of Intensity Images
title_sort	application of second-order histograms in noise reduction and contrast measurement of intensity images
granting_institution	Multimedia University
granting_department	Faculty of Engineering & Technology
publishDate	2019
_version_	1747829677717716992

Application Of Second-Order Histograms In Noise Reduction And Contrast Measurement Of Intensity Images

Similar Items