Experimental evaluation of neural, statistical and model-based approaches to FLIR ATR

Baoxin Li, Qinfen Zheng, Sandor Der, Rama Chellappa, Nasser M. Nasrabadi, Lipchen A. Chan, LinCheng C. Wang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

This paper presents an empirical evaluation of a number of recently developed Automatic Target Recognition Algorithms for Forward-Looking InfraRed(FLIR) imagery using a large database of real second-generation FLIR images. The algorithms evaluated are based on convolution neural networks (CNN), principal component analysis (PCA), linear discriminant analysis (LDA), learning vector quantization (LVQ), and modular neural networks (MNN). Two model-based algorithms, using Hausdorff metric based matching and geometric hashing, are also evaluated. A hierarchical pose estimation system using CNN plus either PCA or LDA, developed by the authors, is also evaluated using the same data set.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
EditorsF.A. Sadjadi
Pages388-397
Number of pages10
Volume3371
DOIs
StatePublished - 1998
Externally publishedYes
EventAutomatic Target Recognition VIII - Orlando, FL, United States
Duration: Apr 13 1998Apr 17 1998

Other

OtherAutomatic Target Recognition VIII
CountryUnited States
CityOrlando, FL
Period4/13/984/17/98

Fingerprint

Discriminant analysis
principal components analysis
Infrared radiation
Neural networks
Convolution
convolution integrals
Principal component analysis
evaluation
infrared imagery
Automatic target recognition
vector quantization
target recognition
Vector quantization
learning

Keywords

  • Automatic target recognition
  • Performance evaluation

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Li, B., Zheng, Q., Der, S., Chellappa, R., Nasrabadi, N. M., Chan, L. A., & Wang, L. C. (1998). Experimental evaluation of neural, statistical and model-based approaches to FLIR ATR. In F. A. Sadjadi (Ed.), Proceedings of SPIE - The International Society for Optical Engineering (Vol. 3371, pp. 388-397) https://doi.org/10.1117/12.323856

Experimental evaluation of neural, statistical and model-based approaches to FLIR ATR. / Li, Baoxin; Zheng, Qinfen; Der, Sandor; Chellappa, Rama; Nasrabadi, Nasser M.; Chan, Lipchen A.; Wang, LinCheng C.

Proceedings of SPIE - The International Society for Optical Engineering. ed. / F.A. Sadjadi. Vol. 3371 1998. p. 388-397.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Li, B, Zheng, Q, Der, S, Chellappa, R, Nasrabadi, NM, Chan, LA & Wang, LC 1998, Experimental evaluation of neural, statistical and model-based approaches to FLIR ATR. in FA Sadjadi (ed.), Proceedings of SPIE - The International Society for Optical Engineering. vol. 3371, pp. 388-397, Automatic Target Recognition VIII, Orlando, FL, United States, 4/13/98. https://doi.org/10.1117/12.323856
Li B, Zheng Q, Der S, Chellappa R, Nasrabadi NM, Chan LA et al. Experimental evaluation of neural, statistical and model-based approaches to FLIR ATR. In Sadjadi FA, editor, Proceedings of SPIE - The International Society for Optical Engineering. Vol. 3371. 1998. p. 388-397 https://doi.org/10.1117/12.323856
Li, Baoxin ; Zheng, Qinfen ; Der, Sandor ; Chellappa, Rama ; Nasrabadi, Nasser M. ; Chan, Lipchen A. ; Wang, LinCheng C. / Experimental evaluation of neural, statistical and model-based approaches to FLIR ATR. Proceedings of SPIE - The International Society for Optical Engineering. editor / F.A. Sadjadi. Vol. 3371 1998. pp. 388-397
@inproceedings{c9f4dbd2f82549e2902ced9fbd9aa4bd,
title = "Experimental evaluation of neural, statistical and model-based approaches to FLIR ATR",
abstract = "This paper presents an empirical evaluation of a number of recently developed Automatic Target Recognition Algorithms for Forward-Looking InfraRed(FLIR) imagery using a large database of real second-generation FLIR images. The algorithms evaluated are based on convolution neural networks (CNN), principal component analysis (PCA), linear discriminant analysis (LDA), learning vector quantization (LVQ), and modular neural networks (MNN). Two model-based algorithms, using Hausdorff metric based matching and geometric hashing, are also evaluated. A hierarchical pose estimation system using CNN plus either PCA or LDA, developed by the authors, is also evaluated using the same data set.",
keywords = "Automatic target recognition, Performance evaluation",
author = "Baoxin Li and Qinfen Zheng and Sandor Der and Rama Chellappa and Nasrabadi, {Nasser M.} and Chan, {Lipchen A.} and Wang, {LinCheng C.}",
year = "1998",
doi = "10.1117/12.323856",
language = "English (US)",
volume = "3371",
pages = "388--397",
editor = "F.A. Sadjadi",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - Experimental evaluation of neural, statistical and model-based approaches to FLIR ATR

AU - Li, Baoxin

AU - Zheng, Qinfen

AU - Der, Sandor

AU - Chellappa, Rama

AU - Nasrabadi, Nasser M.

AU - Chan, Lipchen A.

AU - Wang, LinCheng C.

PY - 1998

Y1 - 1998

N2 - This paper presents an empirical evaluation of a number of recently developed Automatic Target Recognition Algorithms for Forward-Looking InfraRed(FLIR) imagery using a large database of real second-generation FLIR images. The algorithms evaluated are based on convolution neural networks (CNN), principal component analysis (PCA), linear discriminant analysis (LDA), learning vector quantization (LVQ), and modular neural networks (MNN). Two model-based algorithms, using Hausdorff metric based matching and geometric hashing, are also evaluated. A hierarchical pose estimation system using CNN plus either PCA or LDA, developed by the authors, is also evaluated using the same data set.

AB - This paper presents an empirical evaluation of a number of recently developed Automatic Target Recognition Algorithms for Forward-Looking InfraRed(FLIR) imagery using a large database of real second-generation FLIR images. The algorithms evaluated are based on convolution neural networks (CNN), principal component analysis (PCA), linear discriminant analysis (LDA), learning vector quantization (LVQ), and modular neural networks (MNN). Two model-based algorithms, using Hausdorff metric based matching and geometric hashing, are also evaluated. A hierarchical pose estimation system using CNN plus either PCA or LDA, developed by the authors, is also evaluated using the same data set.

KW - Automatic target recognition

KW - Performance evaluation

UR - http://www.scopus.com/inward/record.url?scp=0032404387&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032404387&partnerID=8YFLogxK

U2 - 10.1117/12.323856

DO - 10.1117/12.323856

M3 - Conference contribution

VL - 3371

SP - 388

EP - 397

BT - Proceedings of SPIE - The International Society for Optical Engineering

A2 - Sadjadi, F.A.

ER -