Anomaly detection and target prioritization in planetary imagery via the automated global feature analyzer (AGFA): Progress towards a driver for autonomous C4ISR missions

Wolfgang Fink, Alexander J.W. Brooks, Mark A. Tarbell

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

2 Citations (Scopus)

Abstract

The Automated Global Feature AnalyzerTM (AGFATM) is a generically applicable automated sensor-data-fusion, feature extraction, feature vector clustering, anomaly detection, and target prioritization framework. AGFATM operates in the respective feature space delivered by the sensor(s). In this paper we provide an overview of the inner workings of AGFATMand apply AGFATM to planetary imagery, representative of past, current, and future planetary missions, to demonstrate its automated and objective (i.e., unbiased) anomaly detection and target prioritization (i.e., region-of-interest delineation) capabilities. Imaged operational areas are locally processed via a cascade of image segmentation, visual and geometric feature extraction, agglomerative clustering, and principal components analysis. Resulting clusters are labeled based on relative size and location in feature space. Anomalous regions may be considered immediate targets for follow-up in-situ investigation by local robotic agents, which can be directed via autonomous telecommanding, e.g., as part of a Tier-Scalable Reconnaissance mission architecture. These capabilities will be essential for driving fully autonomous C4ISR missions of the future, since the speed of light prohibits "real time" Earth-controlled conduct of planetary exploration beyond the Moon.

Original languageEnglish (US)
Title of host publicationMicro- and Nanotechnology Sensors, Systems, and Applications X
PublisherSPIE
Volume10639
ISBN (Electronic)9781510617896
DOIs
StatePublished - Jan 1 2018
Event2018 Micro- and Nanotechnology (MNT) Sensors, Systems, and Applications X Conference - Orlando, United States
Duration: Apr 15 2018Apr 19 2018

Other

Other2018 Micro- and Nanotechnology (MNT) Sensors, Systems, and Applications X Conference
CountryUnited States
CityOrlando
Period4/15/184/19/18

Fingerprint

Prioritization
Anomaly Detection
imagery
Driver
Feature extraction
analyzers
anomalies
Feature Space
Light velocity
pattern recognition
Sensor data fusion
Feature Extraction
Target
Moon
Image segmentation
Clustering
Principal component analysis
Sensor Fusion
Robotics
reconnaissance

Keywords

  • Agglomerative clustering
  • Autonomous CISR systems
  • Autonomous decision making
  • Multi-Tiered robotic exploration architectures
  • Objective anomaly detection
  • Principal components analysis
  • Sensor-data-fusion framework
  • Target prioritization

ASJC Scopus subject areas

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

Cite this

Fink, W., Brooks, A. J. W., & Tarbell, M. A. (2018). Anomaly detection and target prioritization in planetary imagery via the automated global feature analyzer (AGFA): Progress towards a driver for autonomous C4ISR missions. In Micro- and Nanotechnology Sensors, Systems, and Applications X (Vol. 10639). [106391Z] SPIE. https://doi.org/10.1117/12.2303795

Anomaly detection and target prioritization in planetary imagery via the automated global feature analyzer (AGFA) : Progress towards a driver for autonomous C4ISR missions. / Fink, Wolfgang; Brooks, Alexander J.W.; Tarbell, Mark A.

Micro- and Nanotechnology Sensors, Systems, and Applications X. Vol. 10639 SPIE, 2018. 106391Z.

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

Fink, W, Brooks, AJW & Tarbell, MA 2018, Anomaly detection and target prioritization in planetary imagery via the automated global feature analyzer (AGFA): Progress towards a driver for autonomous C4ISR missions. in Micro- and Nanotechnology Sensors, Systems, and Applications X. vol. 10639, 106391Z, SPIE, 2018 Micro- and Nanotechnology (MNT) Sensors, Systems, and Applications X Conference, Orlando, United States, 4/15/18. https://doi.org/10.1117/12.2303795
Fink, Wolfgang ; Brooks, Alexander J.W. ; Tarbell, Mark A. / Anomaly detection and target prioritization in planetary imagery via the automated global feature analyzer (AGFA) : Progress towards a driver for autonomous C4ISR missions. Micro- and Nanotechnology Sensors, Systems, and Applications X. Vol. 10639 SPIE, 2018.
@inproceedings{1262895d735546a0805824e0bf8db52d,
title = "Anomaly detection and target prioritization in planetary imagery via the automated global feature analyzer (AGFA): Progress towards a driver for autonomous C4ISR missions",
abstract = "The Automated Global Feature AnalyzerTM (AGFATM) is a generically applicable automated sensor-data-fusion, feature extraction, feature vector clustering, anomaly detection, and target prioritization framework. AGFATM operates in the respective feature space delivered by the sensor(s). In this paper we provide an overview of the inner workings of AGFATMand apply AGFATM to planetary imagery, representative of past, current, and future planetary missions, to demonstrate its automated and objective (i.e., unbiased) anomaly detection and target prioritization (i.e., region-of-interest delineation) capabilities. Imaged operational areas are locally processed via a cascade of image segmentation, visual and geometric feature extraction, agglomerative clustering, and principal components analysis. Resulting clusters are labeled based on relative size and location in feature space. Anomalous regions may be considered immediate targets for follow-up in-situ investigation by local robotic agents, which can be directed via autonomous telecommanding, e.g., as part of a Tier-Scalable Reconnaissance mission architecture. These capabilities will be essential for driving fully autonomous C4ISR missions of the future, since the speed of light prohibits {"}real time{"} Earth-controlled conduct of planetary exploration beyond the Moon.",
keywords = "Agglomerative clustering, Autonomous CISR systems, Autonomous decision making, Multi-Tiered robotic exploration architectures, Objective anomaly detection, Principal components analysis, Sensor-data-fusion framework, Target prioritization",
author = "Wolfgang Fink and Brooks, {Alexander J.W.} and Tarbell, {Mark A.}",
year = "2018",
month = "1",
day = "1",
doi = "10.1117/12.2303795",
language = "English (US)",
volume = "10639",
booktitle = "Micro- and Nanotechnology Sensors, Systems, and Applications X",
publisher = "SPIE",

}

TY - GEN

T1 - Anomaly detection and target prioritization in planetary imagery via the automated global feature analyzer (AGFA)

T2 - Progress towards a driver for autonomous C4ISR missions

AU - Fink, Wolfgang

AU - Brooks, Alexander J.W.

AU - Tarbell, Mark A.

PY - 2018/1/1

Y1 - 2018/1/1

N2 - The Automated Global Feature AnalyzerTM (AGFATM) is a generically applicable automated sensor-data-fusion, feature extraction, feature vector clustering, anomaly detection, and target prioritization framework. AGFATM operates in the respective feature space delivered by the sensor(s). In this paper we provide an overview of the inner workings of AGFATMand apply AGFATM to planetary imagery, representative of past, current, and future planetary missions, to demonstrate its automated and objective (i.e., unbiased) anomaly detection and target prioritization (i.e., region-of-interest delineation) capabilities. Imaged operational areas are locally processed via a cascade of image segmentation, visual and geometric feature extraction, agglomerative clustering, and principal components analysis. Resulting clusters are labeled based on relative size and location in feature space. Anomalous regions may be considered immediate targets for follow-up in-situ investigation by local robotic agents, which can be directed via autonomous telecommanding, e.g., as part of a Tier-Scalable Reconnaissance mission architecture. These capabilities will be essential for driving fully autonomous C4ISR missions of the future, since the speed of light prohibits "real time" Earth-controlled conduct of planetary exploration beyond the Moon.

AB - The Automated Global Feature AnalyzerTM (AGFATM) is a generically applicable automated sensor-data-fusion, feature extraction, feature vector clustering, anomaly detection, and target prioritization framework. AGFATM operates in the respective feature space delivered by the sensor(s). In this paper we provide an overview of the inner workings of AGFATMand apply AGFATM to planetary imagery, representative of past, current, and future planetary missions, to demonstrate its automated and objective (i.e., unbiased) anomaly detection and target prioritization (i.e., region-of-interest delineation) capabilities. Imaged operational areas are locally processed via a cascade of image segmentation, visual and geometric feature extraction, agglomerative clustering, and principal components analysis. Resulting clusters are labeled based on relative size and location in feature space. Anomalous regions may be considered immediate targets for follow-up in-situ investigation by local robotic agents, which can be directed via autonomous telecommanding, e.g., as part of a Tier-Scalable Reconnaissance mission architecture. These capabilities will be essential for driving fully autonomous C4ISR missions of the future, since the speed of light prohibits "real time" Earth-controlled conduct of planetary exploration beyond the Moon.

KW - Agglomerative clustering

KW - Autonomous CISR systems

KW - Autonomous decision making

KW - Multi-Tiered robotic exploration architectures

KW - Objective anomaly detection

KW - Principal components analysis

KW - Sensor-data-fusion framework

KW - Target prioritization

UR - http://www.scopus.com/inward/record.url?scp=85049189340&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85049189340&partnerID=8YFLogxK

U2 - 10.1117/12.2303795

DO - 10.1117/12.2303795

M3 - Conference contribution

AN - SCOPUS:85049189340

VL - 10639

BT - Micro- and Nanotechnology Sensors, Systems, and Applications X

PB - SPIE

ER -