KEYWORDS: Video surveillance, Video, Cameras, Video processing, Imaging systems, Data fusion, Optical character recognition, Lithium, Surveillance, Yield improvement
A new generation of high-resolution surveillance cameras makes it possible to apply video processing and recognition
techniques on live video feeds for the purpose of automatically detecting and identifying objects and events of interest.
This paper addresses a particular application of detecting and identifying vehicles passing through a checkpoint. This
application is of interest to border services agencies and is also related to many other applications. With many
commercial automated License Plate Recognition (LPR) systems available on the market, some of which are available as
a plug-in for surveillance systems, this application still poses many unresolved technological challenges, the main two of
which are: i) multiple and often noisy license plate readings generated for the same vehicle, and ii) failure to detect a
vehicle or license plate altogether when the license plate is occluded or not visible. This paper presents a solution to both
of these problems. A data fusion technique based on the Levenshtein distance is used to resolve the first problem. An
integration of a commercial LPR system with the in-house built Video Analytic Platform is used to solve the latter. The
developed solution has been tested in field environments and has been shown to yield a substantial improvement over
standard off-the-shelf LPR systems.
C-BET is the Comprehensive Biometrics Evaluation Toolkit developed by CBSA in order to analyze
the suitability of biometric systems for fully-automated border/access control applications. Following the multiorder
score analysis and the threshold-validated analysis defined within the C-BET framework, the paper
presents the results of the C-BET evaluation of a commercial voice biometric product. In addition to error
tradeoff and ranking curves traditionally reported elsewhere, the paper presents the results on the newly
introduced performance metrics: threshold-validated recognition ranking and non-confident decisions due to
multiple threshold-validated scores. The results are obtained on over a million voice audio clip comparisons.
Good biometric evaluation practices offered within C-BET framework are presented.
Deploying Video Analytics in operational environments is extremely challenging. This paper presents a methodological
approach developed by the Video Surveillance and Biometrics Section (VSB) of the Science and Engineering Directorate
(S&E) of the Canada Border Services Agency (CBSA) to resolve these problems. A three-phase approach to enable VA
deployment within an operational agency is presented and the Video Analytics Platform and Testbed (VAP/VAT)
developed by the VSB section is introduced. In addition to allowing the integration of third party and in-house built VA
codes into an existing video surveillance infrastructure, VAP/VAT also allows the agency to conduct an unbiased
performance evaluation of the cameras and VA software available on the market. VAP/VAT consists of two
components: EventCapture, which serves to Automatically detect a "Visual Event", and EventBrowser, which serves to
Display & Peruse of "Visual Details" captured at the "Visual Event". To deal with Open architecture as well as with
Closed architecture cameras, two video-feed capture mechanisms have been developed within the EventCapture
component: IPCamCapture and ScreenCapture.
It is not uncommon for contemporary biometric systems to have more than one match below the matching
threshold, or to have two or more matches having close matching scores. This is especially true for those that store large
quantities of identities and/or are applied to measure loosely constrained biometric traits, such as in identification from
video or at a distance. Current biometric performance evaluation standards however are still largely based on measuring
single-score statistics such as False Match, False Non-Match rates and the trade-off curves based thereon. Such
methodology and reporting makes it impossible to investigate the risks and risk mitigation strategies associated with not
having a unique identifying score. To address the issue, Canada Border Services Agency has developed a novel modality-agnostic
multi-order performance analysis framework. The framework allows one to analyze the system performance at
several levels of detail, by defining the traditional single-score-based metrics as Order-1 analysis, and introducing Order-
2 and Order-3 analysis to permit the investigation of the system reliability and the confidence of its recognition decisions.
Implemented in a toolkit called C-BET (Comprehensive Biometrics Evaluation Toolkit), the framework has been applied
in a recent examination of the state-of-the art iris recognition systems, the results of which are presented, and is now
recommended to other agencies interested in testing and tuning the biometric systems.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.