skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Geometry and Gesture-Based Features from Saccadic Eye-Movement as a Biometric in Radiology

Abstract

In this study, we present a novel application of sketch gesture recognition on eye-movement for biometric identification and estimating task expertise. The study was performed for the task of mammographic screening with simultaneous viewing of four coordinated breast views as typically done in clinical practice. Eye-tracking data and diagnostic decisions collected for 100 mammographic cases (25 normal, 25 benign, 50 malignant) and 10 readers (three board certified radiologists and seven radiology residents), formed the corpus for this study. Sketch gesture recognition techniques were employed to extract geometric and gesture-based features from saccadic eye-movements. Our results show that saccadic eye-movement, characterized using sketch-based features, result in more accurate models for predicting individual identity and level of expertise than more traditional eye-tracking features.

Authors:
 [1]; ORCiD logo [2]; ORCiD logo [2]; ORCiD logo [2]
  1. Texas A&M University, College Station
  2. ORNL
Publication Date:
Research Org.:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1394217
DOE Contract Number:
AC05-00OR22725
Resource Type:
Conference
Resource Relation:
Conference: International conference for human computer interaction - Vancouver, , Canada - 7/9/2017 12:00:00 AM-
Country of Publication:
United States
Language:
English

Citation Formats

Hammond, Tracy, Tourassi, Georgia, Yoon, Hong-Jun, and Alamudun, Folami T. Geometry and Gesture-Based Features from Saccadic Eye-Movement as a Biometric in Radiology. United States: N. p., 2017. Web.
Hammond, Tracy, Tourassi, Georgia, Yoon, Hong-Jun, & Alamudun, Folami T. Geometry and Gesture-Based Features from Saccadic Eye-Movement as a Biometric in Radiology. United States.
Hammond, Tracy, Tourassi, Georgia, Yoon, Hong-Jun, and Alamudun, Folami T. Sat . "Geometry and Gesture-Based Features from Saccadic Eye-Movement as a Biometric in Radiology". United States. doi:. https://www.osti.gov/servlets/purl/1394217.
@article{osti_1394217,
title = {Geometry and Gesture-Based Features from Saccadic Eye-Movement as a Biometric in Radiology},
author = {Hammond, Tracy and Tourassi, Georgia and Yoon, Hong-Jun and Alamudun, Folami T.},
abstractNote = {In this study, we present a novel application of sketch gesture recognition on eye-movement for biometric identification and estimating task expertise. The study was performed for the task of mammographic screening with simultaneous viewing of four coordinated breast views as typically done in clinical practice. Eye-tracking data and diagnostic decisions collected for 100 mammographic cases (25 normal, 25 benign, 50 malignant) and 10 readers (three board certified radiologists and seven radiology residents), formed the corpus for this study. Sketch gesture recognition techniques were employed to extract geometric and gesture-based features from saccadic eye-movements. Our results show that saccadic eye-movement, characterized using sketch-based features, result in more accurate models for predicting individual identity and level of expertise than more traditional eye-tracking features.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Sat Jul 01 00:00:00 EDT 2017},
month = {Sat Jul 01 00:00:00 EDT 2017}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • Iris recognition is among the highest accuracy biometrics. However, its accuracy relies on controlled high quality capture data and is negatively affected by several factors such as angle, occlusion, and dilation. Non-ideal iris recognition is a new research focus in biometrics. In this paper, we present a gaze estimation method designed for use in an off-angle iris recognition framework based on the ANONYMIZED biometric eye model. Gaze estimation is an important prerequisite step to correct an off-angle iris images. To achieve the accurate frontal reconstruction of an off-angle iris image, we first need to estimate the eye gaze direction frommore » elliptical features of an iris image. Typically additional information such as well-controlled light sources, head mounted equipment, and multiple cameras are not available. Our approach utilizes only the iris and pupil boundary segmentation allowing it to be applicable to all iris capture hardware. We compare the boundaries with a look-up-table generated by using our biologically inspired biometric eye model and find the closest feature point in the look-up-table to estimate the gaze. Based on the results from real images, the proposed method shows effectiveness in gaze estimation accuracy for our biometric eye model with an average error of approximately 3.5 degrees over a 50 degree range.« less
  • Iris recognition has been proven to be an accurate and reliable biometric. However, the recognition of non-ideal iris images such as off angle images is still an unsolved problem. We propose a new biometric targeted eye model and a method to reconstruct the off-axis eye to its frontal view allowing for recognition using existing methods and algorithms. This allows for existing enterprise level algorithms and approaches to be largely unmodified by using our work as a pre-processor to improve performance. In addition, we describe the `Limbus effect' and its importance for an accurate segmentation of off-axis irides. Our method usesmore » an anatomically accurate human eye model and ray-tracing techniques to compute a transformation function, which reconstructs the iris to its frontal, non-refracted state. Then, the same eye model is used to render a frontal view of the reconstructed iris. The proposed method is fully described and results from synthetic data are shown to establish an upper limit on performance improvement and establish the importance of the proposed approach over traditional linear elliptical unwrapping methods. Our results with synthetic data demonstrate the ability to perform an accurate iris recognition with an image taken as much as 70 degrees off-axis.« less
  • We present results from the optimization of a re-identification process using two sets of biometric data obtained from the Civilian American and European Surface Anthropometry Resource Project (CAESAR) database. The datasets contain real measurements of features for 2378 individuals in a standing (43 features) and seated (16 features) position. A genetic algorithm (GA) was used to search a large combinatorial space where different features are available between the probe (seated) and gallery (standing) datasets. Results show that optimized model predictions obtained using less than half of the 43 gallery features and data from roughly 16% of the individuals available producemore » better re-identification rates than two other approaches that use all the information available.« less
  • This paper reports that the authors developed for the Air Force the Mark VI Personal Identity Verifier (PIV) for controlling access to a fixed or mobile ICBM site, a computer terminal, or mainframe. The Mark VI records the digitized silhouettes of four fingers of each hand on an AT and T smart card. Like fingerprints, finger shapes, lengths, and widths constitute an unguessable biometric password. A Security Officer enrolls an authorized person who places each hand, in turn, on a backlighted panel. An overhead scanning camera records the right and left hand reference templates on the smart card. The Securitymore » Officer adds to the card: name, personal identification number (PIN), and access restrictions such as permitted days of the week, times of day, and doors. To gain access, cardowner inserts card into a reader slot and places either hand on the panel. Resulting access template is matched to the reference template by three sameness algorithms. The final match score is an average of 12 scores (each of the four fingers, matched for shape, length, and width), expressing the degree of sameness. (A perfect match would score 100.00.) The final match score is compared to a predetermined score (threshold), generating an accept or reject decision.« less
  • A new type of biometric identifier which utilizes hand outline measurements made in three dimensions is described. This device uses solid state imaging with no moving parts. The important characteristics of accuracy, speed, user tolerability, small template size, low power, portability and reliability are discussed. A complete stand-alone biometric access control station with sufficient memory for 10,000 users and weighing less than 10 pounds has been built and tested. A test was conducted involving daily use by 112 users over a seven week period during which over 6300 access attempts were made. The single try equal error rate was foundmore » to be 0.4%. There were no false rejects when three tries were allowed before access was denied. Defeat with an artifact is difficult because the hand must be copied in all three dimensions.« less