Masters Theses

Keywords and Phrases

Feature Extraction Toolbox (FET)

Abstract

"Feature Extraction Toolbox (FET) is a generalized pattern-recognition application for analysis and diagnosis of single dimensional multiband signals and two-dimensional multispectral image data. The toolbox can be used in wide variety of engineering applications. FET contains a series of tasks for signal plotting, image processing, image analysis, feature extraction, feature evaluation, training and classification. Initially conceived and developed at Honeywell Federal Manufacturing & Technologies, a second phase of development began at Intelligent System Center, University of Missouri, Rolla .. The goal was to extend the library of image processing, feature-extraction and patternrecognition algorithms. Five new image analysis and pattern classification techniques have been incorporated into the toolbox library. Four of them are related to image processing and analysis and one falls under the pattern recognition system of FET. Apart from this, the Visual Basic Graphical User Interface has been worked upon, to bring forth an ease of use and a systematic approach in performing the tasks.

The later part of the thesis discusses an implementation of a unique image segmentation technique based on non-linear wave propagation to overcome the failure of many other deformable models in applications where fast time response is required. This algorithm was designed to take advantage of parallel processing cellular systems. One such system is the Cellular Neural Network (CNN) paradigm. Parallel array processors based on CNN are ideally suited for image-processing tasks. The active contour model is implemented on CNN Universal Machine simulator for image segmentation. Simulations are carried out on synthetic images and satisfactory results have been reported"--Abstract, page iii.

Advisor(s)

Agarwal, Sanjeev, 1971-

Committee Member(s)

Rao, Vittal S.
Moss, Randy Hays, 1953-

Department(s)

Electrical and Computer Engineering

Degree Name

M.S. in Computer Engineering

Publisher

University of Missouri--Rolla

Publication Date

Summer 2004

Pagination

ix, 66 pages

Note about bibliography

Includes bibliographical references (pages 62-65)

Rights

© 2004 Shiva Kumar Sooryavaram, All rights reserved.

Document Type

Thesis - Restricted Access

File Type

text

Language

English

Subject Headings

Pattern recognition systems Image processing Neural networks (Computer science)

Thesis Number

T 8597

Print OCLC #

58047710

Download
Off-campus S&T users

Share My Thesis If you are the author of this work and would like to grant permission to make it openly accessible to all, please click the button above.

Share

 
COinS