Assessing face acquisition

The requirements necessary for taking a successful face picture are fairly straightforward. The camera must be operational, and the subject must be illuminated sufficiently, facing the camera and in focus. Yet, a significant portion of the facial photographs taken at United States ports of entry are unusable for the purposes of automatic face recognition. In this paper, we consider the usability components of the face image capture process that contribute to the relatively high ratio of unusable images collected by United States Visitor and Immigrant Status Indicator Technology (US-VISIT). In addition, we introduce a general evaluation methodology—including the use of a simple image overlay—to quantify various characteristics of face imagery. The experimental context mimicked the point-of-entry environment, but with specific usability enhancements. The collected data suggests that these usability enhancements may be used to improve face image capture with the current equipment. US-VISIT requested that the biometrics usability team at the National Institute of Standards and Technology (NIST) examine the current US-VISIT face image collection process to identify any usability and human factors that may improve the existing face image capture process. As such this study did not address other technologies or technology solutions. This report presents the results of a study that examined five usability and human factors enhancements to the current US-VISIT collection process: 1. the camera should resemble a traditional camera; 2. the camera should click when the picture is taken to provide feedback to the traveler that the picture is being taken; 3. the camera should be used in portrait mode; 4. the operator should be facing the traveler and the monitor while positioning the camera and 5. provide some marking on the floor (such as footprints) to indicate to the traveler where to stand for the photograph. The study was conducted as follows: first we visited and observed a representative operational setting (Dulles Airport) in order to understand the primary users and the context of use. Based on these observations we identified the 5 usability and human factors enhancements enumerated above that may improve the face image capture process. A usability study was designed that mimicked the operational process but incorporated the 5 enhancements and face images were collected from 300 participants. A visual inspection evaluation methodology based on an image overlay was used to quantify the various characteristics of face imagery based on the face image standards. Results from the visual inspection process compared …