System for characterizing biometric station metrics with genericized biometric information
Inventors
Assignees
US Department of Homeland Security
Publication Number
US-11144773-B2
Publication Date
2021-10-12
Expiration Date
2038-10-03
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
Systems and methods are described, and an example system creates a transaction between a biometric station and a user in a group of users. Transaction includes establishing for the user, a ground-truth identity, capturing an image of the user in the biometric station, making a first determination of whether the image is sufficient for identification of the user. When the image is sufficient, the transaction derives biometric information from the image, converts the biometric information into a generic-form biometric information, agnostic of how the biometric information was obtained, and generated hashed user data by applying a hash function to the generic-form biometric information. The transaction makes a second determination, based on the hashed user data, of whether the user is represented in a gallery of sample images. The transactions are repeated for other users in the group. In the repeats, when the image is determined sufficient, another second determination is made, generating a plurality of second determinations. An acquisition quotient is calculated for the plurality of second determinations, based on one of efficiency, effectiveness, and accuracy relative to time.
Core Innovation
The invention provides systems and methods to quantify biometric acquisition and identification in a controlled environment. An example system creates transactions between a biometric station and users in a user group, establishing ground-truth identity for each user, capturing images at the biometric station, determining whether the images are sufficient for identification, deriving biometric information from sufficient images, converting that biometric information into a generic form agnostic to how it was obtained, generating hashed user data from the generic biometric information, and determining whether the user is represented in a gallery of sample images. These actions are repeated for multiple users to generate a plurality of identification determinations and calculate an acquisition quotient based on efficiency, effectiveness, and accuracy relative to time.
The system includes establishing timestamps to measure transaction times between a user's entry and exit from the biometric station, thereby quantifying interaction durations. The biometric station can perform onboard matching using a provided image gallery or pass biometric information to a backend test facility for analysis. The test facility supports large-scale, real-world statistical testing and evaluation of biometric stations, analyzing efficiency, effectiveness, and user satisfaction over demographically diverse user groups.
The problem addressed arises from the limitations of biometric technologies, including inappropriate form factor, slow acquisition speed, poor user satisfaction, failure to acquire or extract biometric information, and failures in matching. Although biometric matching failures have decreased due to improvements, failures to acquire sufficient biometric data remain challenging since non-collected images typically do not provide data for further analysis. The invention addresses this gap by quantifying acquisition and identification with genericized biometric information and using repeated transactions and statistical analysis to evaluate system performance.
Claims Coverage
The patent includes two independent claims detailing a backend system and a method for characterizing a biometric station, focusing on biometric acquisition, processing, and performance evaluation.
Transaction creation and biometric measurement
Creating transactions between a biometric station and users involving establishing ground-truth identity, generating entry and exit timestamps, capturing user images, assessing image sufficiency for identification, deriving biometric information, genericizing biometric data, hashing the generic data, and determining user presence in a gallery of sample images.
Repeated transaction generation and acquisition quotient calculation
Repeating the transaction process for multiple users to generate multiple identification determinations, determining average transaction times for users, and calculating an acquisition quotient based on one or more of efficiency, effectiveness, and accuracy relative to time.
Generic form biometric information and hashing
Converting biometric information into generic-form biometric information agnostic to the means of acquisition, and applying hash functions to this generic data to generate hashed user data usable for identification comparison.
Handling image sufficiency and failure to acquire metrics
Determining image sufficiency for identification, repeating image capture attempts until sufficiency or time interval expiration, designating failure to acquire when applicable, and calculating failure to acquire rates as a performance metric.
Sorting transactions and calculating biometric effectiveness metrics
Sorting transactions into subsets based on image capture times relative to timestamps, and calculating respective biometric effectiveness metrics for these subsets to evaluate performance.
Ground truth identity establishment via scanning
Establishing ground-truth identity of users by scanning identification tags worn by the users at the biometric station, serving as a definitive linkage for biometric comparison and validation.
The claims cover a backend system and corresponding methods that implement controlled biometric data acquisition transactions, generic biometric data processing, repeated evaluations over user groups, and statistical computations of efficiency and effectiveness metrics to characterize performance of biometric stations.
Stated Advantages
Provides a controlled environment for large-scale biometric station testing enabling statistical analysis of efficiency, effectiveness, and user acceptance.
Supports evaluation of biometric acquisition speed and accuracy relative to time through quantified transaction timestamps and calculation of acquisition quotient.
Utilizes genericized biometric information to be agnostic to capture methods and algorithms, enabling more universal performance characterization.
Incorporates user satisfaction measurements via post-interaction surveys to evaluate user experience comprehensively.
Documented Applications
Testing and evaluating biometric stations within controlled test facilities simulating operational environments such as security checkpoints.
Large-scale biometric system performance measurement for high-throughput identification of travelers and high-security access control systems.
Assessment of biometric systems using a diverse population to support operational suitability and improvement of biometric acquisition and identification technologies.
Use of statistical test designs including subject randomized block design and Latin square counterbalancing for unbiased comparison among multiple biometric stations.
Interested in licensing this patent?