Best Practices in Reporting Time Duration in Biometrics

BEST PRACTICES IN
REPORTING TIME DURATION
IN BIOMETRICS
Ben Petry, Dr. Stephen Elliott, Dr. Richard Guest, Dr.
Mathias Sutton, and Kevin O’Connor

• What is Time?
• Why Examine Time in Biometrics?
• A Short History of Time
• Time in Biometrics
• The General Biometric Model and HBSI Model
• Changes to the General Biometric Model and HBSI Model
• Biometrics Duration Scale Model
• Conclusions
OVERVIEW

• “The measured or measurable period during which an action, process, or condition
exists or continues” [1].
• “Indefinite, unlimited duration in which things are considered as happening in the
past, present, or future; every moment there has ever been or ever will be” [2].
• “The science of arranging time in fixed periods for the purpose of dating events
accurately and arranging them in order of occurrence” [2].
• International Electrotechnical Commission Time Definition (60050-113-01-03): “One-
dimensional subspace of space-time, which is locally orthogonal to space.”
WHAT IS TIME?

• The current model of time in the biometrics field are not consistently applied, as there
are limited definitions that can be used.
• Researchers speak of very specific intervals of data collection which may be
confusing and create a disconnect between studies, making them hard to compare.
• Time, though, “is basically a human construct to fit the needs of humans as we grow
and evolve, it stands to reason that we can and should rethink and try to adapt our
ideas and use of time into something that will be more useful” [3].
• Namely, a new construct should be developed to fit the needs of the field.
WHY EXAMINE TIME IN BIOMETRICS?

• Examples of “time” in biometrics literature:
• “One set in particular, the US-VISIT Point of Entry dataset (POE) contains all
fingerprints collected by US-VISIT between January and June 2004” [15].
• “All of the iris images used in this study were acquired with the same LG 2200
iris imaging camera, located in the same studio throughout the four years of
image acquisition” [16].
• “The database comprises 264,645 iris images of 676 unique subjects captured
over 27 sessions” [17].
• “The final dataset was collected in four acquisition sessions, which spanned a
total of 12 weeks” [18].
WHY EXAMINE TIME IN BIOMETRICS

•Animal bones dating to 20,000 years ago used
to keep track of the number of days between
moon cycles [4].
•2776 BC, Egyptians calculated the year to be
365 days year [4].
A SHORT HISTORY OF TIME

•The Sumerians into divided their days into 12 equal
parts [4].
•1000 years later, the Babylonians created the
modern framework of 24 hour days with each hour
divided into 60 minutes and each minute divided
into 60 seconds [4].

• Romans merged the framework of the Egyptian
system of 365 days per year with the Babylonian
concept of hour, minutes, and seconds [5].
• Developed the Julian calendar system.
• The Gregorian calendar (1582) updated the Julian
calendar to take into account that the Earth takes 365
¼ days to revolve the sun [5].

• Clocks that changed the world
• Clepsydras (water clocks) developed in Egypt (1500 BC) [4]
• Pendulum regulated clock created by Christiaan Huygens
(1656) [6]
• The Harrison chronometer accurate to one second loss per
day and required for accurate Atlantic ocean navigation
(1737) [7]

• Clocks that changed the world
• The most capable mechanical clock is the Shortt clock capable of one
tenth of a second error per day (1921) [8]
• The first atomic clocks developed capable of 0.00001 seconds of error
per day and is more accurate then any measure of time that could be
calculated from astronomical observations (1949) [7]
• Current cesium based clocks are accurate to one trillionth of a second
uncertainty per day [7]

• Coordinated Universal Time (UTC):
• The primary time standard the world sets their clocks to.
• The standard time for the internet, aviation, and other
industries which operate on a global scale with precision.
• Interchangeable with Greenwich Mean Time in most cases.
• Acts as the global ‘midnight’
• Denoted as ‘Zulu’ time (Used by the United States military)
• Does not vary with seasons (Daylight Savings Time)
A SHORT HISTORY OF TIME – WHERE
ARE WE NOW?

A SHORT HISTORY OF TIME – UTC
http://commons.wikimedia.org/wiki/File:World_Time_Zones_Map.png

• ISO 8601 created in 1988, in part to restore “numerous
misconceptions of dates and time”.
• [hh] refers to a zero-padded hour 00 – 24
• [mm] refers to a zero-padded minute 00 – 59
• [ss] refers to a zero-padded second 00 – 60
• Time Zone Designators: Z
• Example: 15:43:07.159Z = 15 hours, 43 minutes, 7.159 seconds
A SHORT HISTORY OF TIME – ISO TIME

• P = the duration designator
(historically called "period")
placed at the start of the
duration representation
• Y = the year designator
• M = the month designator
• W = the week designator
• D = the day designator
• T = the time designator that
precedes the time components
of the representation
• H = the hour designator
• M = the minute designator
• S = the second designator
(NOTING DURATION)
Example: P4Y2M1W4DT2H14M43S
Duration = 4 years, 2 months, 1 week, 4 days, 2 hours, 14 minutes, 43 seconds

•Start and End: 2015-03-24T18:27:30.000Z/2015-03-14T08:00:00.000Z
•Start and Duration: 2015-03-24T18:27:30.000Z/P4Y2M1W4DT2H14M43S
(NOTING TIME INTERVALS)

• IEC 60050-113 (2011)
• Time (113-01-03): “One-dimensional subspace of space time,
which is locally orthogonal to space.”
• Process (113-01-06): “Sequence in time of interrelated events.”
• Time Axis (113-01-07): “Mathematical representation of the
succession in time of instantaneous events along a unique axis.
• Instant (113-01-08): “Point on the time axis.”
A SHORT HISTORY OF TIME – IEC TIME
http://www.electropedia.org/iev/iev.nsf/index?openform&part=113

• IEC 60050-113 (2011)
• Time Interval (113-01-10): “Part of the time axis limited by two instants.”
• Time Scale (113-01-11): “System of ordered marks which can be
attributed to instants of the time axis, one instant being chosen as the
origin.”
• Date (113-01-12): “Mark attributed to an instant by means of a specified
time scale.”
• Duration (113-01-13): “Range of a time interval.”

• IEC 60050-113 (2011)
• Accumulated Duration (113-01-14): “Sum of durations
characterized by given conditions over a given time interval.”
• Calendar Date (113-01-16): “Date on a time scale consisting of a
calendar and a succession of calendar days.”
• Clock Time (113-01-18): “Quantitative expression marking an
instant within a calendar day by the duration elapsed after
midnight in the local standard time.”

•“Time” is purely conceptual. It doesn’t matter if you
divide a day into six parts, 12 parts, or 24 parts, as
year as everyone agrees what those divisions are.
•Time resolution is as accurate as the current
technology allows. The smaller the unit of time, the
more we can understand our surroundings.
A SHORT HISTORY OF TIME –
TAKEAWAYS

• With as many definitions of “time” and associated terminology, there are
just as many subjective measurements:
• Morning, afternoon, or evening have many meanings within the context to time
depending on with whom you ask.
• Additionally, the subdivisions of time may be necessary depending on the
research being conducted.
• Single day a week/month/year collections
• Morning vs evening
• Before/after treatment
TIME IN BIOMETRICS – SUBJECTIVE
MEASUREMENTS

• Definitional
• Operational Times [9]
• The Relationship Between Presentations, Attempts and Transactions [11]
• Non-definitional
• ‘Subject biometric information was collected once every __ (days, weeks,
months, years, semesters, etc.) in __ visits over __ amount of time.’
• Very sporadic time frames that are confusing and difficult to
interpret/replicate.
TIME IN BIOMETRICS – SUBJECTIVE
MEASUREMENTS

•‘Time’ aligns with the specific function the user
or system are undergoing for a duration of
measurable units (seconds for example).
•Helpful for analyzing system throughput
performance.
TIME IN BIOMETRICS – OPERATIONAL
TIMES

• Total Transaction Time: The “…sum of all the subcomponent periods of time
associated with the biometric application system” [9].
• Overt Biometric Transaction Time: “This begins with the biometric sample
presentation and ends with the biometric decision. Therefore, this includes the
presentation of the biometric trait portion of the subject interaction time, biometric
subsystem processing time, which includes sample acquisition and sample
processing time, and the biometric decision time” [9].
• Subject Interaction Time: “…commences when a claim of identity is made (or
presented)… The time ends when the individual has presented his/her biometric
characteristic(s) and the sensor begins to acquire the sample” [9].
TIMES

• Biometric Subsystem Processing/Transaction Time: “…the time taken for the system
to acquire the biometric sample, to evaluate the quality of the sample, and if the
quality is satisfied, to process that sample for comparison. For the samples of bad
quality, the biometric system requests the subject to submit the biometric trait. The
biometric subsystem processing time ends when either a comparison score or a
request for re-submission is generated” [9].
• Biometric Decision Time: “…the time required by the biometric subsystem to
generate an accept or reject response based on the comparison score and the
decision logic” [9].
• External Operation Time: “…the time required to complete the application
transaction” [9].
TIMES

TIMES [9]

• Using the logic of the operational times, [10] found mean
enrollment and mean verification times of hand geometry
recognition machines.
• By measuring and analyzing the frames of collected video
recordings of subject interactions with the machine, the
researchers we able to accurately conclude transaction
times to the closest one-fifteenth of a second.
• “This paper has shown that videos can be automatically
coded post-hoc to determine transaction times without the
use of a human operator” [10].
TIMES APPLIED

•All subject-led
•Physical determination of when the interaction
starts – “interaction volume” – which is
modality and sensor-led
•Use beam-breakers for example
TIME IN BIOMETRICS – WHEN DOES
THE INTERACTION START?

TIME IN BIOMETRICS – THE RELATIONSHIP
BETWEEN PRESENTATIONS, ATTEMPTS AND
TRANSACTIONS [11]

•This framework, at its core, is very ingenious.
•Definitionally, it is quite confusing:
• Are erroneous presentations and correct
presentations classified as the same thing?
• What if the system does not detect a correct
presentation?
TIME IN BIOMETRICS – THE RELATIONSHIP
BETWEEN PRESENTATIONS, ATTEMPTS AND
TRANSACTIONS

•The general biometric model was in 1998 and aims
to identify “the common structures and parallelisms
between seemingly disparate methodologies” [12].
•The most recent version provides better clarity with
regards to data storage, matching, and decision
making processes
TAKING A STEP BACK – THE GENERAL
BIOMETRIC MODEL

THE GENERAL BIOMETRIC MODEL
ISO/IEC 19795-1

• The Human-Biometric Sensor Interaction (HBSI) model was created “to demonstrate
how metrics from biometrics (sample quality and system performance), ergonomics
(physical and cognitive), and usability (efficiency, effectiveness, and satisfaction)
overlap and can be used to evaluate overall functionality and performance of a
biometric system” [13].
• Strengths of the HBSI model
• The ability to define how to user interacts with a system
• What errors, especially common ones, are made by the user
• What is causing these errors to occur repeatedly
• How well does the user need to be trained to correctly interact with the system
THE HUMAN BIOMETRIC-SENSOR
(HBSI) MODEL

• Erroneous Presentation: Any presentation, whether made with malicious intent or not, that was not
performed to the specifications of the particular biometric sensor collecting the sample.
• Defective Interaction (DI): “…occurs when a bad presentation is made to the biometric sensor and
is not detected by the system” [13].
• Concealed Interaction (CI): “…occurs when an erroneous presentation is made to the sensor that is
detected by the biometric system, but is not handled or classified correctly as an ‘error’ by the
biometric system” [13].
• False Interaction (FI): “…occurs when a user presents their biometric features to the biometric
system, which are detected by the system and is correctly classified by the system as erroneous
due to a fault or errors that originated from an incorrect action, behavior, or movement executed by
the user” [13]
HBSI COMPONENT DEFINITIONS

• Correct Presentation: Any presentation that was performed within the specifications of the particular
biometric sensor collecting the sample.
• Failure to Detect (FTD): “…the proportion of presentations to the sensor that are observed by test
personnel but are not detected by the biometric sensor” [13]. There are two types of FTD: system
and external factor.
• Failure to Extract (FTX): “…the proportion of samples that are unable to process or extract
biometric features” [13].
• Successfully Acquired Sample (SAS): “…occurs if a correct presentation is detected by the system
and if biometric features are able to be created from the sample. SAS result from presentations
where biometric features are able to be processed from the captured sample, which are then
passed to the biometric matching systems” [13].
HBSI COMPONENT DEFINITIONS

•These two very important, useful models to
date have not been mapped together.
•This is first necessary before the biometric
duration scale model can be explained
SO WHAT?

•(a) Expansion of the
data capture
subsystem:
•Includes the addition of
different capture
technologies
ALIGNMENT OF THE GENERAL BIOMETRIC
MODEL WITH THE HBSI MODEL

• (a) Expansion of the data
capture subsystem:
• Pre-Processing Capture:
• The system performs some
quality or other analysis during
data capture.
• Once correct collection
specifications have been
obtained, the capture device
detects and captures a sample

•(a) Expansion of the data
capture subsystem:
• Instantaneous Capture:
• The system captures
biometric data that is
presented as is to the
system immediately in
secession of presentation,
sensor, and detection.

• (a) Expansion of the data
capture subsystem:
• Continuous Capture:
• The system captures biometric
data that is presented as is to
the system over a certain
collection period.
• Example: Signature recognition
is continually detecting input
from the user to the sensor.

•(a) Expansion of the
data capture
subsystem:
•If any other these
capture processes fails,
it is a DI or FTD error.

• Specification of the reacquire
loop (b):
• This reacquire loop only occurs
if a FTX, FI, or CI error occurs.
• May happen at any point in
segmentation, feature
extraction, or quality control.
• Note: For CI, these should be
re-acquired if detected by test
administrator or system.

•Division of the model
into three specific
sections:
• Data Capture
• Data Processing
• Data Storage/Matching
ALIGNMENT OF THE HBSI MODEL WITH
THE GENERAL BIOMETRIC MODEL

Uses the mapped general biometric model and
HBSI model to create a timeline to explain what
is happening with the user, the system, and the
resulting output of the two.
BIOMETRIC DURATION SCALE MODEL

•Consists of two
major divisions:
•Phases
•Ranges

• Phases:
• Phases occur as part of the general
biometric model and HBSI model.
• The three models (general biometric,
HBSI, and biometric duration scale)
are all mapped together and are
represented as phases.
• The phases ultimately result in the
enroll/match phase which is the
summary of all scores, metrics, and
other items measured.
BIOMETRIC DURATION SCALE MODEL -
PHASES

• This portion begins outside of the general
biometric model. These are the interactions and
processes that the user undergoes before and
during data capture. It is these interactions that
are determined to be "erroneous" or "correct"
presentations in the HBSI model.
• Interactions begin when a “physical space” is
being entered. This “physical space” is modality
dependent and may change with the research.
“Physical space” should be noted as part of the
publication of the research.
• The actions that are categorized from this
section can yield more information on how
subject interactions can effect system
performance.
– PRESENTATION DEFINITION PHASE

The smallest, most discrete unit
of measurable content that is
collected from a sensor. These
units may be collected as
individual figures in discrete
capture systems or as frames in
continuous capture systems.
This phase occurs in the
presentation sub-system of the
general biometric model. HBSI
errors include DI and FTD. If
these errors occur, move back to
start.
– SAMPLE PHASE

The smallest, most
discrete unit of
measurable content
collected in the data
capture sub-system of a
discrete capture system.
One or many figures are
collected in order to
obtain a sample.
– SAMPLE PHASE (FIGURES)

The smallest, most
discrete unit of
measurable content
collected in the data
capture sub-system of a
continuous capture
system. One or many
frames are collected in
order to obtain a
sample.
– SAMPLE PHASE (FRAMES)

One or more samples are collected
until a sample which contains
measurable properties of the
specific modality are obtained.
These properties are subjected to
the three initial processes of the
signal processing sub-section
(segmentation, feature extraction,
and quality control) of the general
biometric model. HBSI errors
include FI and FTP. If these errors
occur, move back to the start.
– PROCESSING PHASE

In the case of system enrollment, the
template created after the signal
processing sub-system is stored in the
data storage sub-system. This template
and associated metrics are stored in
enroll phase of the biometric time model
regardless if a result of enrollment or
failure to enroll (FTE) occurs. If an FTE
occurs, move back to start. If a CI from
the HBSI model occurs, look back
presentation definition phase and
determine error made by subject. If a
SPS from the HBSI model occurs, end
the phase portion of the model and
submit metrics to day range summary.
– ENROLLMENT PHASE

In the case of system matching, either
verification or identification, the metrics
created from the sample after the signal
processing sub-system is sent to the
matching sub-system of the general
biometric model. These metrics are
compared to a stored user located in the
data storage sub-section. If an FTM
occurs, add metrics to the day range
statistical summary and move back to
start. If a CI from the HBSI model
occurs, look back presentation definition
phase and determine error made by
subject. If a SPS from the HBSI model
occurs, end the phase portion of the
model and submit metrics to day range
summary.
– MATCHING PHASE

– PHASES
Phase Color
Presentation
Definition Phase
Sample Phase
Processing Phase
Enrollment Phase
Matching Phase

All ranges occur on a traditional Gregorian time
frame and UTC. These ranges occur in the
traditional terms "days", "weeks", "months", and
"years". Ranges are summaries of the previous
range or, in the case of day range, enroll and
matching phases.
– RANGES

•A summary of all metrics from enrollment, FTE,
match, FTM, and any another other metrics of
interest collected from 0:00.00 UTC to
23:59.99 UTC for a user, system, or both.
•After such time, a new day range begins.
BIOMETRIC TIMES SCALE – DAY
RANGE

match, FTM, and any other metrics of interest
collected from one to seven day range for a user,
system, or both.
•When an eighth day range occurs or a new
Gregorian calendar week begins (starts on
Sunday), a new week range begins.
– WEEK RANGE

• A summary of all metrics from enrollment, FTE, match, FTM,
and any other metric of interest collected from up to four
week range points for a user, system, or both.
• When a fifth week range occurs or a new Gregorian
calendar month begins, a new month range begins.
• Months are variable (Feb has 28 or 29 days depending on
the year while June and July always have 30 and 31
respectively). However, if a start date with the month and
year are included with the duration, exact month duration
can be determined including leap year.
– MONTH RANGE

match, FTM, and any metric of interest collected
from up to twelve year range points for a user,
system, or both.
•When a thirteenth month range occurs or a
Gregorian calendar year begins, a new year range
begins.
– YEAR RANGE

match, FTM, and any other metric of interest
collected over the life span of the user, system,
or both.
– LIFE

• Intermediate Ranges
• If required in data collection, intermediate ranges are
allowed an encouraged
• Examples:
• Hours: Day range is broken into 24 parts each an hour year.
• Quarterly: Year range is broken into four parts each three
months year.
BIOMETRIC DURATION MODEL –
INTERMEDIATE RANGES

• Characteristic
• Detection Error Trade-Off Curve
• Enrollment
• Equal Error Rate
• Failure to Acquire
• Failure to Enroll
• False Accept Rate
• False Match
• False Match Rate
• False Non-Match
• False Non-Match Rate
• False Reject Rate
• False-Negative Identification Error
Rate
• False-Positive Identification Rate
• Feature
• Genuine Match
• Ground Truth
• Histogram
• Identification Rate
• Impostor Match
• Match
• Match Score
• Presentation
• Receiver Operating Characteristic
Curve
• Sample
• Template
• Transaction
• Zoo Plot
• Modality Specific Metrics
• Any Other Metric of Interest
METRICS THAT CAN BE EXAMINED IN
BIOMETRIC DURATION SCALE MODEL [14]

BIOMETRIC DURATION SCALE MODEL – EXAMPLE WHOLEPresentation Definition
Phase
Sample Phase Processing Phase Enroll Phase Match Phase Day Range Week Range Month Range Year Range Life
Presentation (type) FTD
DD/MM//Y
YYY
Statistical Summary of
one month range
one year range
Statistical Summary
of life of user or
system
one day range
one week range
Presentation (type) Sample FTX
Presentation (type) DI
Presentation (type) Sample SPS FTE
Presentation (type) Sample SPS Enroll
Presentation (type) Sample SPS NA FTM
Presentation (type) Sample SPS NA Match
Presentation (type) FTD DD/MM//YYYY
Presentation (type) Sample CI
one day range
Presentation (type) DI DD/MM//YYYY
one day range
Presentation (type) Sample SPS NA FTM DD/MM//YYYY
one week range
one day range
Presentation (type)
Presentation (type) Sample FI DD/MM/YYYY
Presentation (type) FTD Statistical Summary of
one day range
one week rangePresentation (type) Sample SPS NA Match
Presentation (type) FTD DD/MM/YYYY
one week range
one month range
Presentation (type) DI Statistical Summary of
one day rangePresentation (type) Sample SPS NA Match
one week range
one month range
one year range
one day range

BIOMETRIC DURATION SCALE MODEL –
RELATING TO EXISTED MODEL IN SCIENCE
• Eon: A major division of geological
time, subdivided into eras.
• Era: A major division of time that is
a subdivision of an eon and is itself
subdivided into periods.
• Period: A major division of
geological time; an era is usually
divided into two or more periods.
• Epoch: Any of several divisions of a
geologic period during which a
geologic series is formed. http://www.geomore.com/geologic-time-scale/

Presentation
Definition Phase
Summary of one
week range
Summary of one
month range
Summary of one
year range
Summary of life
of user or system
Summary of one
day range
Presentation (type) Sample SPS Enroll "Scores"
– EXAMPLE STEP 1

– EXAMPLE STEP 2Presentation Definition
Phase
DD/MM//YY
YY
one week range
one month range
one year range
life of user or system
one day range
Presentation (type) Sample SPS FTE
Presentation (type) Sample SPS Enroll
one day range

– EXAMPLE STEP 2
Presentation
Definition Phase
Summary of one
week range
Summary of one
month range
Summary of one year
range
Summary of life of
user or system
Summary of one day
range
Summary of one day
range

Presentation
Definition Phase
Summary of one
month range
Summary of one
year range
Summary of life
of user or system
Summary of one
day range
Summary of one
week range
Summary of one
day range
Summary of one
day range
– EXAMPLE STEP 3

– EXAMPLE STEP 3Presentation
Definition Phase
Summary of one
month range
Summary of one
year range
Summary of life
of user or system
Summary of one
day range
Summary of one
week range
Summary of one
day range
Summary of one
day range

Presentation Definition
Phase
Summary of one month
range
Summary of one year
range
Summary of life of user
or system
Summary of one day
range
Summary of one week
range
Summary of one day
rangePresentation (type) DI
Summary of one day
range
Summary of one week
range
Summary of one day
range
Presentation (type)
– EXAMPLE STEP 4

– EXAMPLE STEP 4
Phase
range
Summary of one year
range
or system
Summary of one day
range
Summary of one week
range
Summary of one day
Summary of one day
range
Summary of one week
range
Summary of one day
range
Presentation (type)

Phase
Summary of one
month range
Summary of one year
range
Summary of life of
user or system
Summary of one day
range
Summary of one week
range
Summary of one day
Summary of one day
range
Summary of one week
range
Summary of one day
range
Presentation (type)
Presentation (type) FTD Summary of one day
range
Summary of one week
rangePresentation (type) Sample SPS NA Match
– EXAMPLE STEP 5

– EXAMPLE STEP 5Presentation Definition
Phase
range
Summary of one year
range
or system
Summary of one day
range
Summary of one week
range
Summary of one day
Summary of one day
range
Summary of one week
range
Summary of one day
range
Presentation (type)
range
Summary of one week

Phase
range
Summary of one year
range
or system
Summary of one day
range
Summary of one week
range
Summary of one day
Summary of one day
range
Summary of one week
range
Summary of one day
range
Presentation (type)
range
Summary of one week
Summary of one week
range
rangePresentation (type) DI Summary of one day
– EXAMPLE STEP 6

Phase
range
Summary of one year
range
or system
Summary of one day
range
Summary of one week
range
Summary of one day
Summary of one day
range
Summary of one week
range
Summary of one day
range
Presentation (type)
range
Summary of one week
Summary of one week
range
range
Presentation (type) DI Summary of one day
– EXAMPLE STEP 6

– EXAMPLE STEP 7
Phase
range
Summary of one year
range
or system
Summary of one day
range
Summary of one week
range
Summary of one day
range
Summary of one day
range
Summary of one week
range
Summary of one day
range
Presentation (type)
range
Summary of one week
Summary of one week
range
range
Presentation (type) DI Summary of one day
Summary of one week
range
range
Summary of one year
range
Summary of one day
range

Phase
range
Summary of one year
range
Summary of life of
user or system
Summary of one day
range
Summary of one week
range
Summary of one day
Summary of one day
range
Summary of one week
range
Summary of one day
range
Presentation (type)
range
Summary of one week
Summary of one week
range
rangePresentation (type) DI Summary of one day
Summary of one week
range
range
Summary of one year
range
Summary of one day
– EXAMPLE STEP 7

• Adoption of ISO 8601 methodology of reporting time and
duration:
• Start and Duration: 2015-03-03T18:18:30.000Z/P4Y2M1W4DT2H14M43S
• Reporting of specific days, weeks, months, and years, in the
Biometric Duration Scale.
• The following common reporting methodology is in
compliance with ISO 21920.
– COMMON REPORTING

•Common reporting sentence structuring should
mimic the following:
• Data collection began on 28 March 2015 and lasted for 4 years, 0 months, 1 week, and 4
days (2015-03-28T18:18:30.000Z/P4Y0M1W4DT2H14M43S). There were seven visits
which occurred in monthly intervals. The time scope of interest for this report is in the
month range. The collection period of interest for this analysis began on 1 April 2015 and
lasted for 30 days (2015-04-01T13:19:30.000Z/P0Y1M0W0DT2H14M43S).

• Data collection began on DAY MONTH YEAR and lasted for Y years, M months, W week,
and D days (yyyy-mm-ddThh:mm:ss.sssZ/PYYMMWWDDTHHHMMMSSS). There
were __ visits which occurred in ____ intervals. The time scope of interest for this report
is in the ____ range. The collection period of interest for this analysis began on DAY
MONTH YEAR and lasted for Y years, M months, W week, and D days (yyyy-mm-
ddThh:mm:ss.sssZ/PYYMMWWDDTHHHMMMSSS).

• ‘Data collection began on ___ ____ ____ and lasted for _ years, _ months, _ week, and _
days (____-__-__T__:__:__.Z/P_Y_M_W_DT__H__M__S). There were __ visits which
occurred in ___ intervals. The time scope of interest for this report is in the ____ range.
The collection period of interest for this analysis began on ___ ___ ___ and lasted for _
years, _ months, _ week, and _ days (____-__-
__T__:__:__.___Z/P_Y_M_W_DT__H__M__S).

• Time is a human construct that should be manipulated as a tool.
• Specifying intervals of time of interest for research is critical.
• The creation of a standardized framework to collect and describe
data is imperative.
• By incorporating preexisting nomenclature from internationally
recognized standards institutes, implementation will be quicker and
easier.
CONCLUSIONS

•With small theoretical changes, the general
biometric model and HBSI model can be mapped
to each other as well as a new biometric time
duration model.
•These models can be used to help explain what
are the downstream and upstream effects of one
model on another.
CONCLUSIONS

• The reporting of specific time metrics is important for the
comparison of different research projects and replicating past
research.
• By utilizing a common vernacular for reporting study duration,
connections may become more apparent.
• Ultimately, it comes down to transparency of the collected data,
declaring the scope of the research, and expressing the findings of
your study in a common vernacular.
CONCLUSIONS

• [1] “Time.” [Online]. Available: http://www.merriam-webster.com/dictionary/time. [Accessed:
28-Jan-2015].
• [2] “Time.” [Online]. Available: http://www.yourdictionary.com/time. [Accessed: 28-Jan-
2015].
• [3] F. Weil, “The Meaning of Time,” 2013. [Online]. Available:
http://www.huffingtonpost.com/frank-a-weil/the-meaning-of-time_b_4351464.html.
[Accessed: 28-Jan-2015].
• [4] J. O’Connor and E. Robertson, “A History of Time: Classical Time,” School of
Mathematics and Statistics University of St. Andrews, Scotland, 2002. [Online]. Available:
http://www-history.mcs.st-and.ac.uk/HistTopics/Time_1.html.
• [5] G. Moyer, “The Gregorian Calendar,” Sci. Am., vol. 246, no. 5, pp. 144–152, May 1982.
• [6] A. Emmerson, “Things are Seldom what they Seem - Christiaan Huygens, the
Pendulum, and the Cycloid,” Horol. Sci. Newsletter., pp. 2–32, 2001.
• [7] S. A. Diddams, J. C. Bergquist, S. R. Jefferts, and C. W. Oates, “Standards of Time and
Frequency at the Outset of the 21st Century,” vol. 306, no. November, pp. 1318–1324,
2004.
• [8] E. Bullard and H. Jolly, “Gravity Measurements in Great Britian,” Geophysics. J. Int., vol.
3, no. 9, pp. 443–477, 1936.
• [9] S. J. Elliott, E. P. Kakula, and R. T. Lazarick, “Operational Times,” in Encyclopedia of
Biometrics, Springer US, 2009, pp. 1022–1025.
• [10] M. E. Brockly and S. J. Elliott, “Automatic Detection of Biometric Transaction Times,” IT
Ind., vol. 1, no. 1, pp. 1–5, 2013.
• [11] “Text of FCD 19795-2, Biometric Performance Testing and Reporting - Part 2: Testing
Methodologies for Technology and Scenario Evaluation,” ISO/IEC JTC 1/SC 37, pp. 1–42,
2006.
• [12] J. L. Wayman, “A Generalized Biometric Identification System Model,” Signals, Syst.
amp; Computer. 1997, Conf. Rec. Thirty-First Asilomar Conf., vol. 1, pp. 291–295, 1998.
• [13] S. J. Elliott, D. Phd, and E. P. Kukula, “A Definitional Framework for the Human-
Biometric Sensor Interaction Model,” pp. 1–6, 2009.
• [14] T. Dunstone and N. Yager, “Definitions,” in Biometric System and Data Analysis, T.
Dunstone and N. Yager, Eds. Springer, 2009, pp. 99–108.
• [15] Hicklin, A., Watson, C., & Ulery, B. “The Myth of Goats : How many people have
fingerprints that are hard to match?” US Department of Commerce, National Institute of
Standards and Technology, pp. 1–24, 2005.
• [16] Baker, S. E., Bowyer, K. W., Flynn, P. J., & Phillips, P. J. “Template Aging in Iris
Biometrics : Evidence of Increased False Reject Rate in ICE 2006.” In Handbook of Iris
Recognition, Springer, pp. 205–218, 2013.
• [17] Arora, S. S., Vatsa, M., & Singh, R. “On Iris Camera Interoperability” In Biometrics:
Theory, Applications, and Systems (BTAS), 2012 IEEE Fifth International Conference on,
pp. 346-352, 2012.
• [18] Connaughton, R., Sgroi, A., Bowyer, K., & Flynn, P. J. “A Multialgorithm Analysis of
Three Iris Biometric Sensors”, Information Forensics and Security, IEEE Transactions on,
vol. 7, no. 3, pp. 919–931, 2012.
REFERENCES

Best Practices in Reporting Time Duration in Biometrics

Recommended

Recommended

More Related Content

Viewers also liked

Viewers also liked (8)

Similar to Best Practices in Reporting Time Duration in Biometrics

Similar to Best Practices in Reporting Time Duration in Biometrics (20)

More from International Center for Biometric Research

More from International Center for Biometric Research (20)

Recently uploaded

Recently uploaded (20)

Best Practices in Reporting Time Duration in Biometrics