DocumentCode :
1211976
Title :
Attributed relational tree approach to signal classification
Author :
Fisher, M.H. ; Ritchings, R.T.
Author_Institution :
Div. of Electron., Coventry Univ., UK
Volume :
141
Issue :
6
fYear :
1994
fDate :
12/1/1994 12:00:00 AM
Firstpage :
319
Lastpage :
324
Abstract :
An automatic method for target identification which uses a pattern recognition algorithm to analyse an ensemble of range image profiles is presented. Such profiles are typical of those produced by high resolution radar and sonar systems employing pulse compression techniques. The approach uses an attributed relational tree model to characterise features extracted from the waveform image profile. The algorithm is capable of learning generic models for each type of target during a supervised training session. Targets are then classified by matching tree models to a database of stored prototypes using a dynamic programming alignment algorithm. Probability attributes are used to model the large amount of scan to scan distortion in the signal caused by target motion. An experimental system has been implemented, and results derived from real data show that a high classification rate can be achieved
Keywords :
dynamic programming; feature extraction; image classification; object recognition; pulse compression; radar imaging; radar target recognition; sonar imaging; sonar target recognition; trees (mathematics); attributed relational tree approach; attributed relational tree model; database; dynamic programming alignment algorithm; generic models; learning; pattern recognition algorithm; probability attributes; pulse compression techniques; radar; range image profiles; scan to scan distortion; signal classification; sonar; target identification; target motion; waveform image profile;
fLanguage :
English
Journal_Title :
Radar, Sonar and Navigation, IEE Proceedings -
Publisher :
iet
ISSN :
1350-2395
Type :
jour
DOI :
10.1049/ip-rsn:19941522
Filename :
338840
Link To Document :
بازگشت