Abstract
Existing techniques for hyperspectral image (HSI) anomaly detection are computationally intensive precluding real-time implementation. The high dimensionality of the spatial/spectral hypercube with associated correlations between spectral bands present significant impediments to real time full hypercube processing that accurately encapsulates the underlying modeling. Traditional techniques have imposed Gaussian models, but these have suffered from significant computational requirements to compute large inverse covariance matrices as well as modeling inaccuracies. We have developed a novel data-driven, non-parametric HSI anomaly detector that has significantly reduced computational complexity with enhanced HSI modeling, providing the capability for real time performance with detection rates that match or surpass existing approaches. This detector, based on the Support Vector Data Description (SVDD), provides accurate, automated modeling of multi-modal data, facilitating effective application of a global background estimation technique which provides the capability for real time operation on a standard PC platform. We have demonstrated one second processing time on hypercubes of dimension 256×256×145, along with superior detection performance relative to alternate detectors. Computation performance analysis has been quantified via processing runtimes on a PC platform, and detection/false-alarm performance is described via Region Operating Characteristic (ROC) curve analysis for the SVDD anomaly detector vs. alternate anomaly detectors.
| Original language | English (US) |
|---|---|
| Title of host publication | Automatic Target Recognition XVIII |
| Volume | 6967 |
| DOIs | |
| State | Published - Jun 17 2008 |
| Event | Automatic Target Recognition XVIII - Orlando, FL, United States Duration: Mar 19 2008 → Mar 20 2008 |
Other
| Other | Automatic Target Recognition XVIII |
|---|---|
| Country | United States |
| City | Orlando, FL |
| Period | 3/19/08 → 3/20/08 |
Fingerprint
Keywords
- Anomaly detection
- HSI
- Hyperspectral
- Real-time anomaly detection
- SVDD
- SVM
ASJC Scopus subject areas
- Electronic, Optical and Magnetic Materials
- Condensed Matter Physics
- Computer Science Applications
- Applied Mathematics
- Electrical and Electronic Engineering
Cite this
Rapid, high performance hyperspectral anomaly detection via global support vector data description. / Meth, Reuven; Banerjee, Amit; Burlina, Philippe; Strat, Thomas.
Automatic Target Recognition XVIII. Vol. 6967 2008. 69670A.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Rapid, high performance hyperspectral anomaly detection via global support vector data description
AU - Meth, Reuven
AU - Banerjee, Amit
AU - Burlina, Philippe
AU - Strat, Thomas
PY - 2008/6/17
Y1 - 2008/6/17
N2 - Existing techniques for hyperspectral image (HSI) anomaly detection are computationally intensive precluding real-time implementation. The high dimensionality of the spatial/spectral hypercube with associated correlations between spectral bands present significant impediments to real time full hypercube processing that accurately encapsulates the underlying modeling. Traditional techniques have imposed Gaussian models, but these have suffered from significant computational requirements to compute large inverse covariance matrices as well as modeling inaccuracies. We have developed a novel data-driven, non-parametric HSI anomaly detector that has significantly reduced computational complexity with enhanced HSI modeling, providing the capability for real time performance with detection rates that match or surpass existing approaches. This detector, based on the Support Vector Data Description (SVDD), provides accurate, automated modeling of multi-modal data, facilitating effective application of a global background estimation technique which provides the capability for real time operation on a standard PC platform. We have demonstrated one second processing time on hypercubes of dimension 256×256×145, along with superior detection performance relative to alternate detectors. Computation performance analysis has been quantified via processing runtimes on a PC platform, and detection/false-alarm performance is described via Region Operating Characteristic (ROC) curve analysis for the SVDD anomaly detector vs. alternate anomaly detectors.
AB - Existing techniques for hyperspectral image (HSI) anomaly detection are computationally intensive precluding real-time implementation. The high dimensionality of the spatial/spectral hypercube with associated correlations between spectral bands present significant impediments to real time full hypercube processing that accurately encapsulates the underlying modeling. Traditional techniques have imposed Gaussian models, but these have suffered from significant computational requirements to compute large inverse covariance matrices as well as modeling inaccuracies. We have developed a novel data-driven, non-parametric HSI anomaly detector that has significantly reduced computational complexity with enhanced HSI modeling, providing the capability for real time performance with detection rates that match or surpass existing approaches. This detector, based on the Support Vector Data Description (SVDD), provides accurate, automated modeling of multi-modal data, facilitating effective application of a global background estimation technique which provides the capability for real time operation on a standard PC platform. We have demonstrated one second processing time on hypercubes of dimension 256×256×145, along with superior detection performance relative to alternate detectors. Computation performance analysis has been quantified via processing runtimes on a PC platform, and detection/false-alarm performance is described via Region Operating Characteristic (ROC) curve analysis for the SVDD anomaly detector vs. alternate anomaly detectors.
KW - Anomaly detection
KW - HSI
KW - Hyperspectral
KW - Real-time anomaly detection
KW - SVDD
KW - SVM
UR - http://www.scopus.com/inward/record.url?scp=44949100470&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=44949100470&partnerID=8YFLogxK
U2 - 10.1117/12.777164
DO - 10.1117/12.777164
M3 - Conference contribution
AN - SCOPUS:44949100470
SN - 9780819471581
VL - 6967
BT - Automatic Target Recognition XVIII
ER -