Block-wise motion detection using compressive imaging system

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

A block-wise motion detection strategy based on compressive imaging, also referred to as feature-specific imaging (FSI), is described in this work. A mixture of Gaussian distributions is used to model the background in a scene. Motion is detected in individual object blocks using feature measurements. Gabor, Hadamard binary and random binary features are studied. Performance of motion detection methods using pixel-wise measurements is analyzed and serves as a baseline for comparison with motion detection techniques based on compressive imaging. ROC (Receiver Operation Characteristic) curves and AUC (Area Under Curve) metrics are used to quantify the algorithm performance. Because a FSI system yields a larger measurement SNR (Signal-to-Noise Ratio) than a traditional system, motion detection methods based on the FSI system have better performance. We show that motion detection algorithms using Hadamard and random binary features in a FSI system yields AUC values of 0.978 and 0.969 respectively. The pixel-based methods are only able to achieve a lower AUC value of 0.627.

Original languageEnglish (US)
Pages (from-to)1170-1180
Number of pages11
JournalOptics Communications
Volume284
Issue number5
DOIs
StatePublished - Mar 1 2011

Fingerprint

Imaging systems
Imaging techniques
Pixels
Gaussian distribution
curves
Signal to noise ratio
pixels
normal density functions
signal to noise ratios
receivers

Keywords

  • Compressive imaging Feature-specific imaging Motion detection Tracking Gaussian mixture model

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Electrical and Electronic Engineering
  • Atomic and Molecular Physics, and Optics
  • Physical and Theoretical Chemistry

Cite this

Block-wise motion detection using compressive imaging system. / Ke, Jun; Ashok, Amit; Neifeld, Mark A.

In: Optics Communications, Vol. 284, No. 5, 01.03.2011, p. 1170-1180.

Research output: Contribution to journalArticle

@article{b5aa686446f1421fbea9ab79d11d4866,
title = "Block-wise motion detection using compressive imaging system",
abstract = "A block-wise motion detection strategy based on compressive imaging, also referred to as feature-specific imaging (FSI), is described in this work. A mixture of Gaussian distributions is used to model the background in a scene. Motion is detected in individual object blocks using feature measurements. Gabor, Hadamard binary and random binary features are studied. Performance of motion detection methods using pixel-wise measurements is analyzed and serves as a baseline for comparison with motion detection techniques based on compressive imaging. ROC (Receiver Operation Characteristic) curves and AUC (Area Under Curve) metrics are used to quantify the algorithm performance. Because a FSI system yields a larger measurement SNR (Signal-to-Noise Ratio) than a traditional system, motion detection methods based on the FSI system have better performance. We show that motion detection algorithms using Hadamard and random binary features in a FSI system yields AUC values of 0.978 and 0.969 respectively. The pixel-based methods are only able to achieve a lower AUC value of 0.627.",
keywords = "Compressive imaging Feature-specific imaging Motion detection Tracking Gaussian mixture model",
author = "Jun Ke and Amit Ashok and Neifeld, {Mark A}",
year = "2011",
month = "3",
day = "1",
doi = "10.1016/j.optcom.2010.11.028",
language = "English (US)",
volume = "284",
pages = "1170--1180",
journal = "Optics Communications",
issn = "0030-4018",
publisher = "Elsevier",
number = "5",

}

TY - JOUR

T1 - Block-wise motion detection using compressive imaging system

AU - Ke, Jun

AU - Ashok, Amit

AU - Neifeld, Mark A

PY - 2011/3/1

Y1 - 2011/3/1

N2 - A block-wise motion detection strategy based on compressive imaging, also referred to as feature-specific imaging (FSI), is described in this work. A mixture of Gaussian distributions is used to model the background in a scene. Motion is detected in individual object blocks using feature measurements. Gabor, Hadamard binary and random binary features are studied. Performance of motion detection methods using pixel-wise measurements is analyzed and serves as a baseline for comparison with motion detection techniques based on compressive imaging. ROC (Receiver Operation Characteristic) curves and AUC (Area Under Curve) metrics are used to quantify the algorithm performance. Because a FSI system yields a larger measurement SNR (Signal-to-Noise Ratio) than a traditional system, motion detection methods based on the FSI system have better performance. We show that motion detection algorithms using Hadamard and random binary features in a FSI system yields AUC values of 0.978 and 0.969 respectively. The pixel-based methods are only able to achieve a lower AUC value of 0.627.

AB - A block-wise motion detection strategy based on compressive imaging, also referred to as feature-specific imaging (FSI), is described in this work. A mixture of Gaussian distributions is used to model the background in a scene. Motion is detected in individual object blocks using feature measurements. Gabor, Hadamard binary and random binary features are studied. Performance of motion detection methods using pixel-wise measurements is analyzed and serves as a baseline for comparison with motion detection techniques based on compressive imaging. ROC (Receiver Operation Characteristic) curves and AUC (Area Under Curve) metrics are used to quantify the algorithm performance. Because a FSI system yields a larger measurement SNR (Signal-to-Noise Ratio) than a traditional system, motion detection methods based on the FSI system have better performance. We show that motion detection algorithms using Hadamard and random binary features in a FSI system yields AUC values of 0.978 and 0.969 respectively. The pixel-based methods are only able to achieve a lower AUC value of 0.627.

KW - Compressive imaging Feature-specific imaging Motion detection Tracking Gaussian mixture model

UR - http://www.scopus.com/inward/record.url?scp=78751641943&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78751641943&partnerID=8YFLogxK

U2 - 10.1016/j.optcom.2010.11.028

DO - 10.1016/j.optcom.2010.11.028

M3 - Article

AN - SCOPUS:78751641943

VL - 284

SP - 1170

EP - 1180

JO - Optics Communications

JF - Optics Communications

SN - 0030-4018

IS - 5

ER -