Publications
Tagged As
Detect-and-avoid closed-loop evaluation of noncooperative well clear definitions
July 12, 2020
Journal Article
Published in:
J. Air Transp., Vol. 28, No. 4, 12 July 2020, pp. 195-206.
Topic:
R&D area:
R&D group:
Summary
Four candidate detect-and-avoid well clear definitions for unmanned aircraft systems encountering noncooperative aircraft are evaluated using safety and operational suitability metrics. These candidates were proposed in previous research based on unmitigated collision risk, maneuver initiation ranges, and other considerations. Noncooperative aircraft refer to aircraft without a functioning transponder. One million encounters representative of the assumed operational environment for the detect-and-avoid system are simulated using a benchmark alerting and guidance algorithm as well as a pilot response model. Results demonstrate sensitivity of the safety metrics to the unmanned aircraft’s speed and the detect-and-avoid system's surveillance volume. The only candidate without a horizontal time threshold, named modified tau, outperforms the other three candidates in avoiding losses of detect and avoid well clear. Furthermore, this candidate's alerting timeline lowers the required surveillance range. This can help reduce the barrier of enabling unmanned aircraft systems' operations with low size, weight, and power sensors.
Summary
Four candidate detect-and-avoid well clear definitions for unmanned aircraft systems encountering noncooperative aircraft are evaluated using safety and operational suitability metrics. These candidates were proposed in previous research based on unmitigated collision risk, maneuver initiation ranges, and other considerations. Noncooperative aircraft refer to aircraft without a functioning transponder. One million...
READ MORE
Augmented Annotation Phase 3
March 9, 2020
Technical Report
Published in:
MIT Lincoln Laboratory Report TR-1248
Summary
Automated visual object detection is an important capability in reducing the burden on human operators in many DoD applications. To train modern deep learning algorithms to recognize desired objects, the algorithms must be "fed" more than 1000 labeled images (for 55%–85% accuracy according to project Maven - Oct 2017 O6, Working Group slide 27) of each particular object. The task of labeling training data for use in machine learning algorithms is human intensive, requires special software, and takes a great deal of time. Estimates from ImageNet, a widely used and publicly available visual object detection dataset, indicate that humans generated four annotations per minute in the overall production of ImageNet annotations. DoD's need is to reduce direct object-by-object human labeling particularly in the video domain where data quantity can be significant. The Augmented Annotations System addresses this need by leveraging a small amount of human annotation effort to propagate human initiated annotations through video to build an initial labeled dataset for training an object detector, and utilizing an automated object detector in an iterative loop to assist humans in pre-annotating new datasets.
Summary
Automated visual object detection is an important capability in reducing the burden on human operators in many DoD applications. To train modern deep learning algorithms to recognize desired objects, the algorithms must be "fed" more than 1000 labeled images (for 55%–85% accuracy according to project Maven - Oct 2017 O6...
READ MORE
Identification and detection of human trafficking using language models
November 26, 2019
Conference Paper
Published in:
European Intelligence and Security Informatics Conf., EISIC, 26-27 November 2019.
Summary
In this paper, we present a novel language model-based method for detecting both human trafficking ads and trafficking indicators. The proposed system leverages language models to learn language structures in adult service ads, automatically select a list of keyword features, and train a machine learning model to detect human trafficking ads. The method is interpretable and adaptable to changing keywords used by traffickers. We apply this method to the Trafficking-10k dataset and show that it achieves better results than the previous models that leverage both ad text and images for detection. Furthermore, we demonstrate that our system can be successfully applied to detect suspected human trafficking organizations and rank these organizations based on their risk scores. This method provides a powerful new capability for law enforcement to rapidly identify ads and organizations that are suspected of human trafficking and allow more proactive policing using data.
Summary
In this paper, we present a novel language model-based method for detecting both human trafficking ads and trafficking indicators. The proposed system leverages language models to learn language structures in adult service ads, automatically select a list of keyword features, and train a machine learning model to detect human trafficking...
READ MORE
Uncovering human trafficking networks through text analysis
October 1, 2019
Conference Paper
Published in:
2019 Grace Hopper Celebration, 1-4 October 2019.
Topic:
R&D area:
Summary
Human trafficking is a form of modern-day slavery affecting an estimated 40 million victims worldwide, primarily through the commercial sexual exploitation of women and children. In the last decade, the advertising of victims has moved from the streets to websites on the Internet, providing greater efficiency and anonymity for sex traffickers. This shift has allowed traffickers to list their victims in multiple geographic areas simultaneously, while also improving operational security by using multiple methods of electronic communication with buyers; complicating the ability of law enforcement to disrupt these illicit organizations. In this presentation, we present a novel unsupervised template matching algorithm for analyzing and detecting complex organizations operating on adult service websites. We apply this method to a large corpus of adult service advertisements retrieved from backpage.com, and show that the networks identified through the algorithm match well with surrogate truth data derived from phone number networks in the same corpus. Further exploration of the results show that the proposed method provides deeper insights into the complex structures of sex trafficking organizations, not possible through networks derived from phone numbers alone. This method provides a powerful new capability for law enforcement to more completely identify and gather evidence about trafficking operations
Summary
Human trafficking is a form of modern-day slavery affecting an estimated 40 million victims worldwide, primarily through the commercial sexual exploitation of women and children. In the last decade, the advertising of victims has moved from the streets to websites on the Internet, providing greater efficiency and anonymity for sex...
READ MORE
XLab: early indications & warning from open source data with application to biological threat
January 3, 2018
Conference Paper
Published in:
Proc. 51st Hawaii Int. Conf. on System Sciences, HICSS 2018, pp. 944-953.
Summary
XLab is an early warning system that addresses a broad range of national security threats using a flexible, rapidly reconfigurable architecture. XLab enables intelligence analysts to visualize, explore, and query a knowledge base constructed from multiple data sources, guided by subject matter expertise codified in threat model graphs. This paper describes a novel system prototype that addresses threats arising from biological weapons of mass destruction. The prototype applies knowledge extraction analytics—including link estimation, entity disambiguation, and event detection—to build a knowledge base of 40 million entities and 140 million relationships from open sources. Exact and inexact subgraph matching analytics enable analysts to search the knowledge base for instances of modeled threats. The paper introduces new methods for inexact matching that accommodate threat models with temporal and geospatial patterns. System performance is demonstrated using several simplified threat models and an embedded scenario.
Summary
XLab is an early warning system that addresses a broad range of national security threats using a flexible, rapidly reconfigurable architecture. XLab enables intelligence analysts to visualize, explore, and query a knowledge base constructed from multiple data sources, guided by subject matter expertise codified in threat model graphs. This paper...
READ MORE
Detecting pathogen exposure during the non-symptomatic incubation period using physiological data
November 13, 2017
Journal Article
Published in:
bioRxiv preprint doi: https://doi.org/10.1101/218818, 13 November 2017.
Summary
Early pathogen exposure detection allows better patient care and faster implementation of public health measures (patient isolation, contact tracing). Existing exposure detection most frequently relies on overt clinical symptoms, namely fever, during the infectious prodromal period. We have developed a robust machine learning based method to better detect asymptomatic states during the incubation period using subtle, sub-clinical physiological markers. Starting with highresolution physiological waveform data from non-human primate studies of viral (Ebola, Marburg, Lassa, and Nipah viruses) and bacterial (Y. pestis) exposure, we processed the data to reduce short-term variability and normalize diurnal variations, then provided these to a supervised random forest classification algorithm and post-classifier declaration logic step to reduce false alarms. In most subjects detection is achieved well before the onset of fever; subject cross-validation across exposure studies (varying viruses, exposure routes, animal species, and target dose) lead to 51h mean early detection (at 0.93 area under the receiver-operating characteristic curve [AUCROC]). Evaluating the algorithm against entirely independent datasets for Lassa, Nipah, and Y. pestis exposures un-used in algorithm training and development yields a mean 51h early warning time (at AUCROC=0.95). We discuss which physiological indicators are most informative for early detection and options for extending this capability to limited datasets such as those available from wearable, non-invasive, ECG-based sensors.
Summary
Early pathogen exposure detection allows better patient care and faster implementation of public health measures (patient isolation, contact tracing). Existing exposure detection most frequently relies on overt clinical symptoms, namely fever, during the infectious prodromal period. We have developed a robust machine learning based method to better detect asymptomatic states...
READ MORE
Use of mass spectrometric vapor analysis to improve canine explosive detection efficiency
June 9, 2017
Journal Article
Published in:
Anal. Chem., Vol. 89, 9 June 2017, 6482-90.
Topic:
R&D area:
R&D group:
Summary
Canines remain the gold standard for explosives detection in many situations, and there is an ongoing desire for them to perform at the highest level. This goal requires canine training to be approached similarly to scientific sensor design. Developing a canine training regimen is made challenging by a lack of understanding of the canine's odor environment, which is dynamic and typically contains multiple odorants. Existing methodology assumes that the handler's intention is an adequate surrogate for actual knowledge of the odors cuing the canine, but canines are easily exposed to unintentional explosive odors through training material cross-contamination. A sensitive, real-time (~1 s) vapor analysis mass spectrometer was developed to provide tools, techniques, and knowledge to better understand, train, and utilize canines. The instrument has a detection library of nine explosives and explosive-related materials consisting of 2,4-dinitrotoluene (2,4-DNT), 2,6-dinitrotoluene (2,6-DNT), 2,4,6-trinitrotoluene (TNT), nitroglycerin (NG), 1,3,5-trinitroperhydro-1,3,5-triazine (RDX), pentaerythritol tetranitrate (PETN), triacetone triperoxide (TATP), hexamethylene triperoxide diamine (HMTD), and cyclohexanone, with detection limits in the parts-per-trillion to parts-per-quadrillion range by volume. The instrument can illustrate aspects of vapor plume dynamics, such as detecting plume filaments at a distance. The instrument was deployed to support canine training in the field, detecting cross-contamination among training materials, and developing an evaluation method based on the odor environment. Support for training material production and handling was provided by studying the dynamic headspace of a nonexplosive HMTD training aid that is in development. These results supported existing canine training and identified certain areas that may be improved.
Summary
Canines remain the gold standard for explosives detection in many situations, and there is an ongoing desire for them to perform at the highest level. This goal requires canine training to be approached similarly to scientific sensor design. Developing a canine training regimen is made challenging by a lack of...
READ MORE
Aircraft laser strike geolocation system
June 5, 2017
Conference Paper
Published in:
17th AIAA Aviation Technology, Integration, and Operations Conf., 5-9 June 2017.
Topic:
R&D area:
Summary
Laser strikes against aircraft are increasing at an alarming rate, driven by the availability of cheap powerful lasers and a lack of deterrence due to the challenges of locating and apprehending perpetrators. Although window coatings and pilot goggles effectively block laser light, uptake has been low due to high cost and pilot reluctance. This paper describes the development and testing of a proof-of-concept ground based sensor system to rapidly geolocate the origin of a laser beam in a protected region of airspace and disseminate this information to law enforcement to allow a timely and targeted response. Geolocation estimates with accuracies of better than 20 m have been demonstrated within 30 seconds of an event at a range of 8.9 nmi with a 450 mW laser. Recommendations for an operational prototype at an airport are also described.
Summary
Laser strikes against aircraft are increasing at an alarming rate, driven by the availability of cheap powerful lasers and a lack of deterrence due to the challenges of locating and apprehending perpetrators. Although window coatings and pilot goggles effectively block laser light, uptake has been low due to high cost...
READ MORE
Germanium CCDs for large-format SWIR and x-ray imaging
May 25, 2017
Journal Article
Published in:
J. Instrum., Vol. 12, No. 5, May 2017, C05014.
Topic:
R&D area:
R&D group:
Summary
Germanium exhibits high sensitivity to short-wave infrared (SWIR) and X-ray radiation, making it an interesting candidate for imaging applications in these bands. Recent advances in germanium processing allow for high-quality charge-coupled devices (CCDs) to be realized in this material. In this article, we discuss our evaluation of germanium as an absorber material for CCDs via fabrication and analysis of discrete devices such as diodes, metal-insulator-semiconductor capacitors, and buried-channel metal-oxide-semiconductor field-effect transistors (MOSFETs). We then describe fabrication of our first imaging device on germanium, a 32 x 1 x 8.1 um linear shift register. Based on this work, we find that germanium is a promising material for CCDs imaging in the SWIR and X-ray bands.
Summary
Germanium exhibits high sensitivity to short-wave infrared (SWIR) and X-ray radiation, making it an interesting candidate for imaging applications in these bands. Recent advances in germanium processing allow for high-quality charge-coupled devices (CCDs) to be realized in this material. In this article, we discuss our evaluation of germanium as an...
READ MORE
Detecting virus exposure during the pre-symptomatic incubation period using physiological data
March 3, 2017
Journal Article
Published in:
Science Translational Medicine (submitted), 3 March 2017.
Summary
Early pathogen exposure detection allows better patient care and faster implementation of public health measures (patient isolation, contact tracing). Existing exposure detection most frequently relies on overt clinical symptoms, namely fever, during the infectious prodromal period. We have developed a robust machine learning method to better detect asymptomatic states during the incubation period using subtle, sub-clinical physiological markers. Using high-resolution physiological data from non-human primate studies of Ebola and Marburg viruses, we pre-processed the data to reduce short-term variability and normalize diurnal variations, then provided these to a supervised random forest classification algorithm. In most subjects detection is achieved well before the onset of fever; subject cross-validation lead to 52±14h mean early detection (at >0.90 area under the receiver-operating characteristic curve). Cross-cohort tests across pathogens and exposure routes also lead to successful early detection (28±16h and 43±22h, respectively). We discuss which physiological indicators are most informative for early detection and options for extending this capability to lower data resolution and wearable, non-invasive sensors.
Summary
Early pathogen exposure detection allows better patient care and faster implementation of public health measures (patient isolation, contact tracing). Existing exposure detection most frequently relies on overt clinical symptoms, namely fever, during the infectious prodromal period. We have developed a robust machine learning method to better detect asymptomatic states during...
READ MORE