Exploring EO Vehicle Recognition Performance using Manifolds as a Function of Lighting Condition Variability

Olga Mendoza-Schrock; Mateen M. Rizki; Edmund G. Zelnio; Vincent J. Velten; Frederick D. Garber; Michael L. Raymer; John C. Gallagher

doi:10.1117/12.2178408

Exploring EO Vehicle Recognition Performance using Manifolds as a Function of Lighting Condition Variability

Olga Mendoza-Schrock, Mateen M. Rizki, Edmund G. Zelnio, Vincent J. Velten, Frederick D. Garber, Michael L. Raymer, John C. Gallagher

Computer Science and Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Novel techniques are necessary in order to improve the current state-of-the-art for Aided Target Recognition (AiTR) especially for persistent intelligence, surveillance, and reconnaissance (ISR). A fundamental assumption that current AiTR systems make is that operating conditions remain semi-consistent between the training samples and the testing samples. Today’s electro-optical AiTR systems are still not robust to common occurrences such as changes in lighting conditions. In this work, we explore the effect of systemic variation in lighting conditions on vehicle recognition performance. In addition, we explore the use of low-dimensional nonlinear representations of high-dimensional data derived from electro-optical synthetic vehicle images using Manifold Learning - specifically Diffusion Maps on recognition. Diffusion maps have been shown to be a valuable tool for extraction of the inherent underlying structure in high-dimensional data.

Original language	American English
Title of host publication	Proceedings Volume 9464, Ground/Air Multisensor Interoperability, Integration, and Networking for Persistent ISR VI
Editors	Michael A. Kolodny, Tien Pham
Publisher	SPIE
Volume	9464
ISBN (Electronic)	9781628415803
DOIs	https://doi.org/10.1117/12.2178408
State	Published - May 20 2015
Event	Ground/Air Multisensor Interoperability, Integration, and Networking for Persistent ISR VI - Baltimore, United States Duration: Apr 20 2015 → Apr 22 2015

Conference

Conference	Ground/Air Multisensor Interoperability, Integration, and Networking for Persistent ISR VI
Country/Territory	United States
City	Baltimore
Period	4/20/15 → 4/22/15

ASJC Scopus Subject Areas

Electronic, Optical and Magnetic Materials
Condensed Matter Physics
Computer Science Applications
Applied Mathematics
Electrical and Electronic Engineering

Keywords

Aided Target Recognition
Diffusion Maps
Electro-Imagery
Manifold Learning
Vehicle Classification

Disciplines

Computer Sciences
Engineering

Access to Document

10.1117/12.2178408License: Unspecified

Cite this

Mendoza-Schrock, O., Rizki, M. M., Zelnio, E. G., Velten, V. J., Garber, F. D., Raymer, M. L., & Gallagher, J. C. (2015). Exploring EO Vehicle Recognition Performance using Manifolds as a Function of Lighting Condition Variability. In M. A. Kolodny, & T. Pham (Eds.), Proceedings Volume 9464, Ground/Air Multisensor Interoperability, Integration, and Networking for Persistent ISR VI (Vol. 9464). Article 94640I SPIE. https://doi.org/10.1117/12.2178408

Exploring EO Vehicle Recognition Performance using Manifolds as a Function of Lighting Condition Variability. / Mendoza-Schrock, Olga; Rizki, Mateen M.; Zelnio, Edmund G. et al.
Proceedings Volume 9464, Ground/Air Multisensor Interoperability, Integration, and Networking for Persistent ISR VI. ed. / Michael A. Kolodny; Tien Pham. Vol. 9464 SPIE, 2015. 94640I.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Mendoza-Schrock, O, Rizki, MM, Zelnio, EG, Velten, VJ, Garber, FD, Raymer, ML & Gallagher, JC 2015, Exploring EO Vehicle Recognition Performance using Manifolds as a Function of Lighting Condition Variability. in MA Kolodny & T Pham (eds), Proceedings Volume 9464, Ground/Air Multisensor Interoperability, Integration, and Networking for Persistent ISR VI. vol. 9464, 94640I, SPIE, Ground/Air Multisensor Interoperability, Integration, and Networking for Persistent ISR VI, Baltimore, United States, 4/20/15. https://doi.org/10.1117/12.2178408

Mendoza-Schrock O, Rizki MM, Zelnio EG, Velten VJ, Garber FD, Raymer ML et al. Exploring EO Vehicle Recognition Performance using Manifolds as a Function of Lighting Condition Variability. In Kolodny MA, Pham T, editors, Proceedings Volume 9464, Ground/Air Multisensor Interoperability, Integration, and Networking for Persistent ISR VI. Vol. 9464. SPIE. 2015. 94640I doi: 10.1117/12.2178408

@inproceedings{5ce3b931b9be4d91a13265bba8a629ec,

title = "Exploring EO Vehicle Recognition Performance using Manifolds as a Function of Lighting Condition Variability",

abstract = " Novel techniques are necessary in order to improve the current state-of-the-art for Aided Target Recognition (AiTR) especially for persistent intelligence, surveillance, and reconnaissance (ISR). A fundamental assumption that current AiTR systems make is that operating conditions remain semi-consistent between the training samples and the testing samples. Today{\textquoteright}s electro-optical AiTR systems are still not robust to common occurrences such as changes in lighting conditions. In this work, we explore the effect of systemic variation in lighting conditions on vehicle recognition performance. In addition, we explore the use of low-dimensional nonlinear representations of high-dimensional data derived from electro-optical synthetic vehicle images using Manifold Learning - specifically Diffusion Maps on recognition. Diffusion maps have been shown to be a valuable tool for extraction of the inherent underlying structure in high-dimensional data.",

keywords = "Aided Target Recognition, Diffusion Maps, Electro-Imagery, Manifold Learning, Vehicle Classification",

author = "Olga Mendoza-Schrock and Rizki, {Mateen M.} and Zelnio, {Edmund G.} and Velten, {Vincent J.} and Garber, {Frederick D.} and Raymer, {Michael L.} and Gallagher, {John C.}",

note = "Publisher Copyright: {\textcopyright} 2015 SPIE.; Ground/Air Multisensor Interoperability, Integration, and Networking for Persistent ISR VI ; Conference date: 20-04-2015 Through 22-04-2015",

year = "2015",

month = may,

day = "20",

doi = "10.1117/12.2178408",

language = "American English",

volume = "9464",

editor = "Kolodny, {Michael A.} and Tien Pham",

booktitle = "Proceedings Volume 9464, Ground/Air Multisensor Interoperability, Integration, and Networking for Persistent ISR VI",

publisher = "SPIE",

}

TY - GEN

T1 - Exploring EO Vehicle Recognition Performance using Manifolds as a Function of Lighting Condition Variability

AU - Mendoza-Schrock, Olga

AU - Rizki, Mateen M.

AU - Zelnio, Edmund G.

AU - Velten, Vincent J.

AU - Garber, Frederick D.

AU - Raymer, Michael L.

AU - Gallagher, John C.

PY - 2015/5/20

Y1 - 2015/5/20

N2 - Novel techniques are necessary in order to improve the current state-of-the-art for Aided Target Recognition (AiTR) especially for persistent intelligence, surveillance, and reconnaissance (ISR). A fundamental assumption that current AiTR systems make is that operating conditions remain semi-consistent between the training samples and the testing samples. Today’s electro-optical AiTR systems are still not robust to common occurrences such as changes in lighting conditions. In this work, we explore the effect of systemic variation in lighting conditions on vehicle recognition performance. In addition, we explore the use of low-dimensional nonlinear representations of high-dimensional data derived from electro-optical synthetic vehicle images using Manifold Learning - specifically Diffusion Maps on recognition. Diffusion maps have been shown to be a valuable tool for extraction of the inherent underlying structure in high-dimensional data.

AB - Novel techniques are necessary in order to improve the current state-of-the-art for Aided Target Recognition (AiTR) especially for persistent intelligence, surveillance, and reconnaissance (ISR). A fundamental assumption that current AiTR systems make is that operating conditions remain semi-consistent between the training samples and the testing samples. Today’s electro-optical AiTR systems are still not robust to common occurrences such as changes in lighting conditions. In this work, we explore the effect of systemic variation in lighting conditions on vehicle recognition performance. In addition, we explore the use of low-dimensional nonlinear representations of high-dimensional data derived from electro-optical synthetic vehicle images using Manifold Learning - specifically Diffusion Maps on recognition. Diffusion maps have been shown to be a valuable tool for extraction of the inherent underlying structure in high-dimensional data.

KW - Aided Target Recognition

KW - Diffusion Maps

KW - Electro-Imagery

KW - Manifold Learning

KW - Vehicle Classification

UR - http://www.scopus.com/inward/record.url?scp=84954053840&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84954053840&partnerID=8YFLogxK

UR - https://corescholar.libraries.wright.edu/cse/370

U2 - 10.1117/12.2178408

DO - 10.1117/12.2178408

M3 - Conference contribution

VL - 9464

BT - Proceedings Volume 9464, Ground/Air Multisensor Interoperability, Integration, and Networking for Persistent ISR VI

A2 - Kolodny, Michael A.

A2 - Pham, Tien

PB - SPIE

T2 - Ground/Air Multisensor Interoperability, Integration, and Networking for Persistent ISR VI

Y2 - 20 April 2015 through 22 April 2015

ER -

Exploring EO Vehicle Recognition Performance using Manifolds as a Function of Lighting Condition Variability

Abstract

Conference

ASJC Scopus Subject Areas

Keywords

Disciplines

Access to Document

Other files and links

Cite this