Dr. John B. Sigman Profile

Dr. John B. Sigman

Postdoctoral Researcher at Thayer School of Engineering at Dartmouth

SPIE Involvement:

Author

Publications (15)

Proceedings Article | 26 May 2020 Paper

Background adaptive faster R-CNN for semi-supervised convolutional object detection of threats in x-ray images

John Sigman, Gregory Spell, Kevin Liang, Lawrence Carin

Proceedings Volume 11404, 1140404 (2020) https://doi.org/10.1117/12.2558542

KEYWORDS: X-rays, Data modeling, X-ray imaging, Feature extraction, Transportation security, Convolutional neural networks, Machine learning, Threat warning systems

Read Abstract +

Recently, progress has been made in the supervised training of Convolutional Object Detectors (e.g. Faster RCNN) for threat recognition in carry-on luggage using X-ray images. This is part of the Transportation Security Administration's (TSA's) mission to ensure safety for air travelers in the United States. Collecting more data reliably improves performance for this class of deep algorithm, but requires time and money to produce training data with threats staged in realistic contexts. In contrast to these hand-collected data containing threats, data from the real-world, known as the Stream-of-Commerce (SOC), can be collected quickly with minimal cost; while technically unlabeled, in this work we make a practical assumption that these are without threat objects. Because of these data constraints, we will use both labeled and unlabeled sources of data for the automatic threat recognition problem. In this paper, we present a semi-supervised approach for this problem which we call Background Adaptive Faster R-CNN. This approach is a training method for two-stage object detectors which uses Domain Adaptation methods from the field of deep learning. The data sources described earlier are considered two “domains": one a hand-collected data domain of images with threats, and the other a real-world domain of images assumed without threats. Two domain discriminators, one for discriminating object proposals and one for image features, are adversarially trained to prevent encoding domain-specific information. Penalizing this encoding is important because otherwise the Convolutional Neural Network (CNN) can learn to distinguish images from the two sources based on superficial characteristics, and minimize a purely supervised loss function without improving its ability to recognize objects. For the hand-collected data, only object proposals and image features completely outside of areas corresponding to ground truth object bounding boxes (background) are used. The losses for these domain-adaptive discriminators are added to the Faster R-CNN losses of images from both domains. This technique enables threat recognition based on examples from the labeled data, and can reduce false alarm rates by matching the statistics of extracted features on the hand-collected backgrounds to that of the real world data. Performance improvements are demonstrated on two independently-collected datasets of labeled threats.

Proceedings Article | 8 May 2018 Presentation + Paper

Accounting for the influence of salt water in the physics required for processing underwater UXO EMI signals

Fridon Shubitidze, Benjamin Barrowes, Irma Shamatava , John Sigman, Kevin O'Neill

Proceedings Volume 10628, 106280L (2018) https://doi.org/10.1117/12.2305161

KEYWORDS: Physics, Signal processing, Electromagnetic coupling, Submerged target modeling, Environmental sensing, Electromagnetism, Instrument modeling, Sensors

Read Abstract +

Proceedings Article | 27 April 2018 Presentation + Paper

Automatic threat recognition of prohibited items at aviation checkpoint with x-ray imaging: a deep learning approach

Kevin Liang, Geert Heilmann, Christopher Gregory, Souleymane Diallo, David Carlson, Gregory Spell, John Sigman, Kris Roe, Lawrence Carin

Proceedings Volume 10632, 1063203 (2018) https://doi.org/10.1117/12.2309484

KEYWORDS: Firearms, Data modeling, X-ray imaging, Performance modeling, Detection and tracking algorithms, Prototyping, Transportation security

Read Abstract +

Proceedings Article | 4 May 2017 Paper

Ultra-wide-band EMI sensing for subsurface deplete uranium detection and classification

Fridon Shubitidze, Benjamin Barrowes, John Sigman, Yinlin Wang, Janet Simms, Hollis Bennett, Don Yule, Steven Larson, Kevin O'Neill

Proceedings Volume 10182, 1018216 (2017) https://doi.org/10.1117/12.2262950

KEYWORDS: Uranium, Electromagnetic coupling, Sensors, Defense and security, Aerosols, Combustion, Metals, Contamination, Environmental management, Soil contamination, Data modeling, Magnetism, Soil science, Receivers, Electromagnetism

Read Abstract +

Proceedings Article | 4 May 2017 Presentation + Paper

A hybrid coil system for high frequency electromagnetic induction sensing

John Sigman, Benjamin Barrowes, Yinlin Wang, Hollis Bennett, Janet Simms, Donald Yule, Kevin O'Neill, Fridon Shubitidze

Proceedings Volume 10182, 101820F (2017) https://doi.org/10.1117/12.2263198

KEYWORDS: Electromagnetic coupling, Signal to noise ratio, Electromagnetism, Target detection, Sensors, Optical spheres, Composites, Land mines

Read Abstract +

Proceedings Article | 3 May 2017 Presentation + Paper

Void and landmine detection using the HFEMI sensor

Benjamin Barrowes, Fridon Shubitidze, John Sigman, Jay Bennett, Janet Simms, Don Yule, Kevin O'Neill

Proceedings Volume 10182, 1018210 (2017) https://doi.org/10.1117/12.2262619

KEYWORDS: Data modeling, Data acquisition, Optical spheres, Electromagnetism, Land mines, Carbon, Sensors, Target detection, Electromagnetic coupling

Read Abstract +

Proceedings Article | 19 May 2016 Paper

A high power EMI sensor for detecting and classifying small and deep targets

F. Shubitidze, B. Barrowes, Yinlin Wang, Irma Shamatava, J. Sigman, K. O'Neil, Daniel Steinhurst

Proceedings Volume 9823, 982308 (2016) https://doi.org/10.1117/12.2224407

KEYWORDS: Systems modeling, Sensors, Electromagnetic coupling, Target detection, Transmitters, Electromagnetism, Receivers, Magnetism, Data modeling, Data processing

Read Abstract +

Proceedings Article | 17 May 2016 Paper

Coil design considerations for a high-frequency electromagnetic induction sensing instrument

John Sigman, Benjamin Barrowes, Yinlin Wang, Hollis Bennett, Janet Simms, Donald Yule, Kevin O'Neill, Fridon Shubitidze

Proceedings Volume 9823, 982302 (2016) https://doi.org/10.1117/12.2223988

KEYWORDS: Electromagnetism, Carbon, Signal to noise ratio, Transmitters, Electromagnetic coupling, Data modeling, Sensors, Inductive coupling, Improvised explosive devices, Prototyping

Read Abstract +

Proceedings Article | 17 May 2016 Paper

Advanced EMI models for survey data processing: targets detection and classification

F. Shubitidze, B. Barrowes, Yinlin Wang, Irma Shamatava, J. Sigman, K. O'Neill

Proceedings Volume 9823, 98230O (2016) https://doi.org/10.1117/12.2224420

KEYWORDS: Data modeling, Electromagnetic coupling, Target detection, Magnetism, Electromagnetism, Sensors, Data processing, Polarizability, Detection and tracking algorithms, Associative arrays

Read Abstract +

Proceedings Article | 3 May 2016 Paper

Carbon fiber and void detection using high-frequency electromagnetic induction techniques

Benjamin Barrowes, John Sigman, YinLin Wang, Kevin O'Neill, Fridon Shubitidze, Janet Simms, Hollis Bennett, Donald Yule

Proceedings Volume 9823, 98230D (2016) https://doi.org/10.1117/12.2224584

KEYWORDS: Electromagnetism, Carbon, Electromagnetic coupling, Metals, Magnetism, Sensors, Optical spheres, Diffusion, Dielectrics, Physics

Read Abstract +

Proceedings Article | 9 June 2014 Paper

Advanced EMI models for survey data processing: targets detection and classification

F. Shubitidze, J. Sigman, Yinlin Wang, J. Miller, J. Keranen, I. Shamatava, B. Barrowes, K. O'Neill

Proceedings Volume 9072, 90720J (2014) https://doi.org/10.1117/12.2050897

KEYWORDS: Electromagnetic coupling, Data modeling, Target detection, Magnetism, Data processing, Transmitters, Sensors, Detection and tracking algorithms, Systems modeling, Polarizability

Read Abstract +

One of the most challenging aspects of survey data processing is target selection. The fundamental input for the classification is dynamic data collected along survey lines. These data are different from the static data obtained in cued mode and used for target classification. Survey data are typically collected using just one transmitter loop (the Z-axis loop) and feature short data point collection times and short decay transience. The collection intervals for each data point are typically 0.1 s, and the signal repetition rates are typically 90 or 270 Hz (in other words, the transient decay times are 2.7 ms or 0.9 ms). Reliable classification requires multiple side/angle illumination; i.e., to conduct reliable classification it is necessary to combine and jointly invert multiple data points. However, picking data points that provide optimal information for classifying targets is a difficult task. The traditional method plots signal amplitudes on a 2D map and picks peaks of signal level without properly accounting for the underlying physics. In this paper, the joint diagonalization is applied to survey data sets to improve data pre-processing and target picking. The JD technique is an EMI data analysis and target classification technique and is applicable for all next-generation multi-static array EMI sensors. The method extracts multi-static response data matrix eigenvalues. The eigenvalues are main characteristics of the data. Recent studies have demonstrated that the method has great potential to quickly estimate the number of potential targets and moreover classify these targets at the data pre-processing stage, in real time and without the need for a forward model. Another advantage of JD is that it provides the ability to separate signal from noise making it possible to de-noise data without distorting the signal due to the targets. In this paper the JD technique is used to process dynamic data collected at South West Proving Ground and Aberdeen Proving Ground (APG) sites using the 2 × 2 TEMTADS and OPTEMA systems, respectively. The joint eigenvalues are extracted as functions of time for each data point and summed/stacked together before being used to create detection maps. Once targets are detected, a set of data is chosen for each anomaly and inverted using the ortho-normalized volume magnetic source technique.

Proceedings Article | 9 June 2014 Paper

Detecting and classifying small and deep targets using improved EMI hardware and data processing approach

F. Shubitidze, B. Barrowes, J. Sigman, Yinlin Wang, Irma Shamatava, K. O'Neill

Proceedings Volume 9072, 90720I (2014) https://doi.org/10.1117/12.2050893

KEYWORDS: Electromagnetic coupling, Target detection, Data modeling, Sensors, Transmitters, Signal to noise ratio, Magnetism, Data processing, Polarizability, Interference (communication)

Read Abstract +

Proceedings Article | 9 June 2014 Paper

A combined joint diagonalization-MUSIC algorithm for subsurface targets localization

Yinlin Wang, John Sigman, Benjamin Barrowes, Kevin O'Neill, Fridon Shubitidze

Proceedings Volume 9072, 90720G (2014) https://doi.org/10.1117/12.2050787

KEYWORDS: Interference (communication), Detection and tracking algorithms, Signal to noise ratio, Sensors, Matrices, Electromagnetic coupling, Electromagnetism, Electronic filtering, Detector arrays, Algorithm development

Read Abstract +

Proceedings Article | 9 June 2014 Paper

Automatic classification of unexploded ordnance applied to live sites for MetalMapper sensor

John Brevard Sigman, Kevin O'Neill, Benjamin Barrowes, Yinlin Wang, Fridon Shubitidze

Proceedings Volume 9072, 90720F (2014) https://doi.org/10.1117/12.2050784

KEYWORDS: Sensors, Polarizability, Data modeling, Detection and tracking algorithms, Magnetic sensors, Magnetism, Machine learning, Electromagnetism, Electromagnetic coupling, Transmitters

Read Abstract +

Proceedings Article | 7 June 2013 Paper

Automatic classification of unexploded ordnance applied to Spencer Range live site for 5x5 TEMTADS sensor

John Sigman, Benjamin Barrowes, Kevin O'Neill, Fridon Shubitidze

Proceedings Volume 8709, 870904 (2013) https://doi.org/10.1117/12.2016118

KEYWORDS: Sensors, Magnetism, Detection and tracking algorithms, Data modeling, Electromagnetism, Electromagnetic coupling, Receivers, Transmitters, Polarizability, Weapons

Read Abstract +

Showing 5 of 15 publications

View contact details

UPDATE YOUR PROFILE

Is this your profile? Update it now.

Sign into your SPIE.org account

Don’t have a profile and want one?

Create an account on SPIE.org

Keywords/Phrases

Search In:

Publication Years