Skip to main content
U.S. flag

An official website of the United States government

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

Information Access Division Highlights - 2015

Information Access Division Highlights - 2015

 

NOVEMBER – DECEMBER 2015

SEPTEMBER - OCTOBER 2015

Computer scientist Kevin Mangold, ITL's Information Access Division, received the Next Generation Award from the American National Standards Institute (ANSI) for his significant contributions to national and international standardization activities in biometrics and identity management, as well as ongoing commitment to the industry, the nation, and the enhancement of the global voluntary consensus standards system. The award is presented to outstanding members of ANSI who have been with the association for less than eight years.

JPEG 2000 CODEC Certification Guidance for 1000 ppi Fingerprint Friction Ridge Imagery
By Shahram Orandi, John Libert, Michael Garris, John Grantham, and Fred Byers
NIST Special Publication 500-300
June 2015

This document describes the procedure by which applications of JPEG 2000 CODECs will be evaluated with respect to conformance to the NIST guidance for compression of 1000 ppi (pixels per inch) friction ridge images as detailed in NIST Special Publication (SP) 500-289. It describes the attributes of a set of fingerprint images selected for conformance testing and the rationale for selection of these images based on both examiner assessment of image quality over increasing degrees of JPEG 2000 compression and relative fidelity based on computational metrics described SP 500-289 and supporting studies. The document also provides background behind the conformance testing, describes the CODEC pathways to be tested and the metrics used to measure compliance, and provides instructions on how to run the protocol and submit results to NIST for evaluation.

JULY - AUGUST 2015

ITL Focuses on Automated Image-Based Tattoo Recognition

ITL researchers are working to develop automated methods to recognize tattoos in images, a technology which would provide significant advantages to law enforcement agencies nationwide. To advance this effort, ITL recently hosted the Tattoo Recognition Technology –Challenge (Tatt-C) Workshop, where an international group of experts from industry, academia, and government gathered at NIST to discuss challenges and potential approaches to automated image-based tattoo recognition. The goals of the workshop were to discuss current law enforcement use cases and state-of-the-art algorithm performance against those use cases, examine technical successes and challenges, share utility and perspectives on the operational use of tattoos, identify gaps and needs to support and progress future development, and identify potential follow-on development and evaluation activities.

The six research teams that participated in the Tatt-C program presented their methodologies, current algorithm performance, and recommendations on progressing development. The participating organizations included MorphoTrak (U.S.), Purdue University (U.S.), the MITRE Corporation (U.S.), Compass Technical Consulting (U.S.), the Fraunhofer Institute of Optronics, System Technologies and Image Exploitation (Germany), and the Alternative Energies and Atomic Energy Commission (France).

Additionally, a panel of U.S. government agencies with operational needs for image-based tattoo recognition technology, including the Federal Bureau of Investigation (FBI) Scars, Marks, Tattoos Services Team, the FBI Cryptanalysis and Racketeering Records Unit, the Department of Homeland Security Homeland Security Investigations, the National Center for Missing and Exploited Children, the Michigan State Police, and the NIST lead for the American National Standards Institute Biometric Data format standard shared their perspectives on current shortcomings of the existing keyword-based tattoo collection and retrieval process and the drivers for an image-based capability.

Computer vision methods applied to representing and matching tattoo images were discussed. Sessions on best practices for tattoo collection and annotation, and defining tattoo similarity identified a number of activities that might advance the technology.

The highlights from the Tatt-C workshop were recently published in Nature Magazine. More information about Tatt-C and the proceedings of the workshop are available on the Tatt-C website.

It's About the Face Impostor Distribution
By P. Jonathon Phillips, Amy N. Yates, Geof H. Givens, and J. Ross Beveridge
NISTIR 8051
April 2015

This report presents a study of the effects of factors on the false accept rate (FAR) for three modern video face recognition algorithms. We examined the effects of environment (location), video (imagery)-based, and demographic factors. The study is performed on the handheld video in the Point and Shoot Face Recognition Challenge (PaSC), which consists of 1401 handheld videos of 265 subjects. The results of our analysis are consistent across the three algorithms. Our analysis shows that FAR can significantly vary.

Face Recognition Vendor Test (FRVT) – Performance of Automated Gender Classification Algorithms
By Mei Ngan and Patrick Grother
NISTIR 8052
April 2015

NIST performed a large-scale empirical evaluation of facial gender classification algorithms, with participation from five commercial providers and one university, using large operational datasets comprised of facial images from visas and law enforcement mugshots, leveraging a combined corpus of close to one million images. Using a lights-out, black-box testing methodology, core gender classification accuracy was baselined over a large dataset composed of images collected under well-controlled pose, illumination, and facial expression conditions, then assessed demographically by gender, age group, and ethnicity. Analysis on commonly benchmarked "in the wild" (i.e., unconstrained) datasets was conducted and compared with those from the constrained dataset. Assessments of classification performance on sketches and gender verification accuracy were documented.

 

MAY - JUNE 2015

Tattoo Recognition Technology – Challenge (Tatt-C)
Date: Held June 8, 2015
Place: NIST, Gaithersburg, Maryland
Sponsor: NIST
Cost: None

Launched in September 2014, the Tatt-C activity challenges the commercial and academic community in advancing research and development into automated image-based tattoo matching technology with goals to determine which methodologies are most effective and whether any are viable for identified operational use cases. The workshop will bring together Tatt-C participants from industry and academia and key sponsors and stakeholders to discuss current tattoo detection and matching performance and successes/technical challenges; share utility and perspectives on the operational use of tattoos; identify gaps and needs to support and progress future development; and shape the follow-on evaluation activity.

 

MARCH - APRIL 2015

The Twenty-Third Text REtrieval Conference Proceedings (TREC 2014)
Ellen Voorhees and Angela Ellis, Editors
NIST Special Publication 500-308
February 2014

This report constitutes the proceedings of the Twenty-Third Text REtrieval Conference (TREC 2014) held in Gaithersburg, Maryland, November 19-21, 2014. The conference was co-sponsored by the National Institute of Standards and Technology (NIST) and the Defense Advanced Research Projects Agency (DARPA).

Integrating Electronic Health Records into Clinical Workflow: An Application of Human Factors Modeling Methods to Specialty Care in Obstetrics and Gynecology and Ophthalmology
By Svetlana Lowry, Mala Ramaiah, E.S. Patterson, D. Brick, M.C. Gibbons, and L.A. Paul
NISTIR 8042
February 2015

A human factors workflow modeling tool, process mapping, was used to visualize and document insights and the end-user needs to improve EHR workflow for clinicians in two specialty outpatient care settings: 1) Obstetrics and Gynecology (Ob-Gyn); and 2) Ophthalmology.The report proposes targeted recommendations for EHR developers and Ob-Gyn and Ophthalmology centers to improve workflow integration with EHRs to improve quality of care and patient safety, and to reduce medical-legal exposure.

 

JANUARY - FEBRUARY 2015

Text REtrieval Conference Series Supports Information Retrieval Research Community

Started in 1992 and cosponsored by the Department of Defense, ITL's Text REtrieval Conference (TREC) series of evaluation workshops is designed to support the information retrieval community by providing the infrastructure necessary for large-scale evaluation of text retrieval methodologies. Over the years, the workshops have evolved based on emerging information retrieval technologies. The evaluation effort has grown in both number of participating systems and the number of countries represented.

Each TREC is organized around a set of focus areas called tracks. TREC participants use their own search engines and a common data set to perform a track's task. They submit their search results to NIST, which uses the combined result sets from all participants to build evaluation resources that are then used to score each participant's submission. Both the evaluation resources and the participants' submissions are made publicly available through the TREC website to support the larger retrieval research community.

TREC 2014, the 23rd conference in the series, was held on November 19-21, 2014. The conference included eight tracks that drew 75 participating teams representing 20 countries. The tracks investigated topics ranging from risk minimization in web search to search over microblogs to efficiently monitoring the information associated with an event such as a natural disaster in real time.

TREC 2014 added an exciting new track called the Clinical Decision Support (CDS) track. The goal was to develop systems that can support healthcare providers by finding clinically relevant information within the vast biomedical literature. This initial year of the track used case narratives developed by physicians at the National Library of Medicine as patient surrogates. A case report is a well-formed narrative summarizing portions of a patient's medical record that typically describes a challenging medical case. The literature base was the open-access portion of PubMed Central. The systems' task was to retrieve articles that contained pertinent information for one of three clinically relevant questions for the target patient: What is the diagnosis? What test should be performed? What treatment should be undertaken? Twenty-six teams participated in the CDS track. The retrieval results suggest that the task is challenging, but feasible, for existing search systems. See the TREC website for more information.

Fingerprint Vendor Technology Evaluation
By Craig Watson, Gregory Fiumara, Elham Tabassi, Su Lan Cheng, Patricia Flanagan, and Wayne Salamon
NISTIR 8034
December 2014

FpVTE was conducted primarily to assess the current capabilities of fingerprint matching algorithms using operational datasets containing several million subjects. There were three classes of participation that examined various finger combinations from single finger all the way up to ten fingers. Enrollment sets varied in size from 10,000 subjects up to 5 million subjects. All data used was sequestered operational data that was not shared with any of the participants. The evaluation provided feedback to the participants after the first two of three submissions, allowing participants to evaluate their performance, make adjustments to their algorithms, and resubmit for further testing. The evaluation was conducted at NIST using NIST-owned hardware. Participants submitted software libraries compliant to the testing Application Programming Interface (API), which were linked to a NIST-developed test driver and run by NIST employees. All participant libraries went through validation testing to ensure that results at NIST matched results participants were getting on their hardware. This is the first large scale one-to-many fingerprint evaluation since FpVTE 2003.

Created October 20, 2015, Updated August 25, 2016