Abstract

Automatic Target Recognition (ATR) systems have significant impact in defense applications. There is a continuing need to develop new and robust techniques to handle the increasingly complex ATR problem. The objectives of this thesis are two-fold. First a new technique to be used for ATR is developed and secondly an integrated ATR system to investigate and combine all subsystems is developed. In this thesis, we have developed a new technique for the feature extraction stage of ATR problem using wavelet transforms. Wavelet transforms have been one of the widely investigated areas of research in the past few years. The promising results obtained in this research indicate that target recognition problems lie within the general scope of wavelet transforms.

Computer programs which implement different tasks of the ATR algorithm namely preprocessing, detection, segmentation, feature extraction and classification are developed. The output obtained after each stage of the algorithm is demonstrated and analyzed. The advantages of using wavelet transforms for feature extraction are discussed. An overview of the complete ATR system is presented.

Department(s)

Computer Science

Second Department

Mechanical and Aerospace Engineering

Third Department

Mathematics and Statistics

Comments

The first Author is a Graduate Student.

This report is substantially the M.S. thesis of the first author, completed December 1994.

Report Number

CSc-94-27

Document Type

Technical Report

Document Version

Final Version

File Type

text

Language(s)

English

Rights

© 1994 University of Missouri - Rolla, All rights reserved

Publication Date

1 December, 1994

Download

Share

 
COinS