Scalable information-optimal compressive target recognition

Ronan Kerviche, Amit Ashok

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

Abstract

We present a scalable information-optimal compressive imager optimized for the target classification task, discriminating between two target classes. Compressive projections are optimized using the Cauchy-Schwarz Mutual Information (CSMI) metric, which provides an upper-bound to the probability of error of target classification. The optimized measurements provide significant performance improvement relative to random and PCA secant projections. We validate the simulation performance of information-optimal compressive measurements with experimental data.

Original languageEnglish (US)
Title of host publicationComputational Imaging
PublisherSPIE
Volume9870
ISBN (Electronic)9781510601116
DOIs
StatePublished - 2016
EventComputational Imaging - Baltimore, United States
Duration: Apr 17 2016Apr 18 2016

Other

OtherComputational Imaging
CountryUnited States
CityBaltimore
Period4/17/164/18/16

Fingerprint

Target Recognition
target recognition
Target
projection
Image sensors
Projection
Chord or secant line
Imager
Mutual Information
Cauchy
Experimental Data
Upper bound
Metric
Simulation
simulation

Keywords

  • Cauchy-Schwarz Mutual Information
  • Classification
  • Compressive Imaging
  • Target Recognition

ASJC Scopus subject areas

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

Cite this

Kerviche, R., & Ashok, A. (2016). Scalable information-optimal compressive target recognition. In Computational Imaging (Vol. 9870). [987008] SPIE. https://doi.org/10.1117/12.2228570

Scalable information-optimal compressive target recognition. / Kerviche, Ronan; Ashok, Amit.

Computational Imaging. Vol. 9870 SPIE, 2016. 987008.

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

Kerviche, R & Ashok, A 2016, Scalable information-optimal compressive target recognition. in Computational Imaging. vol. 9870, 987008, SPIE, Computational Imaging, Baltimore, United States, 4/17/16. https://doi.org/10.1117/12.2228570
Kerviche R, Ashok A. Scalable information-optimal compressive target recognition. In Computational Imaging. Vol. 9870. SPIE. 2016. 987008 https://doi.org/10.1117/12.2228570
Kerviche, Ronan ; Ashok, Amit. / Scalable information-optimal compressive target recognition. Computational Imaging. Vol. 9870 SPIE, 2016.
@inproceedings{e1cfa42f25154ad18b6c526ae1d9aaf1,
title = "Scalable information-optimal compressive target recognition",
abstract = "We present a scalable information-optimal compressive imager optimized for the target classification task, discriminating between two target classes. Compressive projections are optimized using the Cauchy-Schwarz Mutual Information (CSMI) metric, which provides an upper-bound to the probability of error of target classification. The optimized measurements provide significant performance improvement relative to random and PCA secant projections. We validate the simulation performance of information-optimal compressive measurements with experimental data.",
keywords = "Cauchy-Schwarz Mutual Information, Classification, Compressive Imaging, Target Recognition",
author = "Ronan Kerviche and Amit Ashok",
year = "2016",
doi = "10.1117/12.2228570",
language = "English (US)",
volume = "9870",
booktitle = "Computational Imaging",
publisher = "SPIE",
address = "United States",

}

TY - GEN

T1 - Scalable information-optimal compressive target recognition

AU - Kerviche, Ronan

AU - Ashok, Amit

PY - 2016

Y1 - 2016

N2 - We present a scalable information-optimal compressive imager optimized for the target classification task, discriminating between two target classes. Compressive projections are optimized using the Cauchy-Schwarz Mutual Information (CSMI) metric, which provides an upper-bound to the probability of error of target classification. The optimized measurements provide significant performance improvement relative to random and PCA secant projections. We validate the simulation performance of information-optimal compressive measurements with experimental data.

AB - We present a scalable information-optimal compressive imager optimized for the target classification task, discriminating between two target classes. Compressive projections are optimized using the Cauchy-Schwarz Mutual Information (CSMI) metric, which provides an upper-bound to the probability of error of target classification. The optimized measurements provide significant performance improvement relative to random and PCA secant projections. We validate the simulation performance of information-optimal compressive measurements with experimental data.

KW - Cauchy-Schwarz Mutual Information

KW - Classification

KW - Compressive Imaging

KW - Target Recognition

UR - http://www.scopus.com/inward/record.url?scp=84991463786&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84991463786&partnerID=8YFLogxK

U2 - 10.1117/12.2228570

DO - 10.1117/12.2228570

M3 - Conference contribution

VL - 9870

BT - Computational Imaging

PB - SPIE

ER -