Dissertation/Thesis Abstract

Comparison of Features and Mask Size in Thermographic Images of Chiari (CLMS/COMS) Dogs for Syrinx Identification
by Pant, Gita, M.S., Southern Illinois University at Edwardsville, 2018, 67; 13420135
Abstract (Summary)

Thermographic imaging techniques produce images whose pixel values represent the temperature distribution of an object. This research focuses on features that can be utilized for the classification of syrinx identification (CLMS/COMS). The thermographic image data from 19° C to 40° C was linearly remapped to create images with 256 gray level values. Features were extracted from these images, including histogram, second-order histogram based texture, and Laws texture features. Various pattern classification methods have performed well in thermographic image classification problems such as k-nearest neighbor and nearest neighbor for classification, softmax scaling and standard normal density for data normalization, and comparison metrics such as Euclidean distance and normalized vector inner product. Experiments with these features and methods are conducted to determine the best feature set and pattern classification algorithm for syrinx identification in canines.

Indexing (document details)
Advisor: Umbaugh, Scott E.
Commitee: LeAnder, Robert, York, Timothy
School: Southern Illinois University at Edwardsville
Department: Electrical Engineering
School Location: United States -- Illinois
Source: MAI 58/04M(E), Masters Abstracts International
Subjects: Engineering, Electrical engineering
Keywords: Chiari dogs, Syrinx, Thermographic images
Publication Number: 13420135
ISBN: 978-0-438-81933-7
Copyright © 2021 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy