Award Date

8-1-2022

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Computer Science

First Committee Member

Kazem Taghva

Second Committee Member

Venkatesan Muthukumar

Third Committee Member

Laxmi Gewali

Fourth Committee Member

Fatma Nasoz

Fifth Committee Member

Mingon Kang

Sixth Committee Member

Emma Regentova

Number of Pages

113

Abstract

The primary object of this dissertation is to investigate the application of hyperspectral technology to accommodate for the growing demand in the automatic dietary assessment applications. Food intake is one of the main factors that contribute to human health. In other words, it is necessary to get information about the amount of nutrition and vitamins that a human body requires through a daily diet. Manual dietary assessments are time-consuming and are also not precise enough, especially when the information is used for the care and treatment of hospitalized patients. Moreover, the data must be analyzed by nutritional experts. Therefore, researchers have developed various semiautomatic or automatic dietary assessment systems; most of them are based on the conventional color images such as RGB. The main disadvantage of such systems is their inability to differentiate foods of similar color or same ingredients in various colors, or different forms such as cooked or mixed forms. Although adding features such as shape, size and texture improve the overall performance, they are sensitive to changes in the illumination, rotation, scale, etc. A balance between quality and quantity of features representation, and system efficiency must also be considered. Hyperspectral technology combines conventional imaging technology with spectroscopy in a three-dimensional data-cube to obtain both the spatial and spectral information of the objects. However, the high dimensionality of hyperspectral data in addition to the redundancy between spectral bands limits performance, especially in online or onboard data processing applications. Thus, various features selection/extraction are also used to select the optimal feature subsets. The results are promising and verify the feasibility of using hyperspectral technology in dietary assessment applications.

Keywords

3D image; Dietary Assessment; Food Recognition; Hyperspectral Imaging; Machine Learning

Disciplines

Artificial Intelligence and Robotics | Computer Engineering | Computer Sciences | Food Science

File Format

pdf

File Size

4900 KB

Degree Grantor

University of Nevada, Las Vegas

Language

English

Rights

IN COPYRIGHT. For more information about this rights statement, please visit http://rightsstatements.org/vocab/InC/1.0/

Download


Share

COinS