U.S. patent application number 16/531488 was filed with the patent office on 2019-11-21 for systems and methods of biometric analysis with adaptive trigger.
This patent application is currently assigned to Princeton Identity, Inc.. The applicant listed for this patent is Princedton Identity, Inc.. Invention is credited to David Alan Ackerman, James R. Bergen, Barry E. Mapen, Steven N. Perna.
Application Number | 20190356827 16/531488 |
Document ID | / |
Family ID | 59961086 |
Filed Date | 2019-11-21 |
United States Patent
Application |
20190356827 |
Kind Code |
A1 |
Ackerman; David Alan ; et
al. |
November 21, 2019 |
SYSTEMS AND METHODS OF BIOMETRIC ANALYSIS WITH ADAPTIVE TRIGGER
Abstract
Exemplary embodiments are directed to biometric analysis systems
including one or more illumination sources configured to provide
dim illumination to a scene including an object and configured to
provide flash illumination to the object in the scene. The
biometric analysis systems include a rolling shutter camera
configured to capture one or more images. The biometric analysis
systems include an adaptive trigger module configured to analyze
the scene to detect the object in the scene during dim illumination
of the scene, determine a position in a frame of the rolling
shutter camera that coincides with the detected object in the
scene, and arrange a delay between a start of image writing by the
rolling shutter camera and a trigger of the one or more
illumination sources such that a stripe of the flash illumination
coincides with the detected object in the scene.
Inventors: |
Ackerman; David Alan;
(Hopewell, NJ) ; Bergen; James R.; (Hopewell,
NJ) ; Mapen; Barry E.; (Stonington, CT) ;
Perna; Steven N.; (Lawrenceville, NJ) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Princedton Identity, Inc. |
Hamilton |
NJ |
US |
|
|
Assignee: |
Princeton Identity, Inc.
Hamilton
NJ
|
Family ID: |
59961086 |
Appl. No.: |
16/531488 |
Filed: |
August 5, 2019 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
15471131 |
Mar 28, 2017 |
10373008 |
|
|
16531488 |
|
|
|
|
62316347 |
Mar 31, 2016 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H04N 5/2256 20130101;
H04N 5/2329 20130101 |
International
Class: |
H04N 5/225 20060101
H04N005/225; H04N 5/232 20060101 H04N005/232 |
Claims
1. A biometric analysis system, comprising: one or more
illumination sources configured to provide dim illumination to a
scene including an object and configured to provide flash
illumination to the object in the scene; a rolling shutter camera
configured to capture one or more images; and an adaptive trigger
module configured to (i) analyze the scene to detect the object in
the scene during dim illumination of the scene, (ii) determine a
position in a frame of the rolling shutter camera that coincides
with the detected object in the scene, and (iii) arrange a delay
between a start of image writing by the rolling shutter camera and
a trigger of the one or more illumination sources such that a
stripe of the flash illumination coincides with the detected object
in the scene.
2. The biometric analysis system of claim 1, wherein the adaptive
trigger module is configured to track movement of the object within
a field-of-view of the rolling shutter camera.
3. The biometric analysis system of claim 2, wherein the adaptive
trigger module is configured to modify the delay between the start
of image writing by the rolling shutter camera and the trigger of
the one or more illumination sources based on detected movement of
the object within the field-of-view.
4. The biometric analysis system of claim 1, wherein the adaptive
trigger module is configured to detect a region of interest of the
object and arranges the delay such that the stripe of flash
illumination coincides with the detected region of interest of the
object.
5. The biometric analysis system of claim 4, wherein the region of
interest of the object includes eyes of a person.
6. The biometric analysis system of claim 1, wherein the object is
a person, and wherein the adaptive trigger module comprises a face
finder configured to detect a face of the person.
7. The biometric analysis system of claim 1, wherein the object is
a physical item, and the adaptive trigger module comprises an
identifier finder configured to detect a unique identifier
associated with the physical item.
8. The biometric analysis system of claim 7, wherein the unique
identifier is a barcode or a quick response (QR) code.
9. The biometric analysis system of claim 1, wherein the one or
more illumination sources are configured to provide the flash
illumination as a synchronized pulse of flash illumination.
10. The biometric analysis system of claim 1, wherein the flash
illumination provided by the one or more illumination sources is
brighter than the dim illumination provided by the one or more
illumination sources.
11. The biometric analysis system of claim 1, wherein the one or
more illumination sources comprise a first illumination source
configured to provide the dim illumination and a second
illumination source configured to provide the flash
illumination.
12. The biometric analysis system of claim 1, wherein the one or
more illumination sources are near infrared illumination
sources.
13. The biometric analysis system of claim 1, wherein the one or
more illumination sources are ambient light.
14. The biometric analysis system of claim 1, wherein the adaptive
trigger module is configured to sweep an illuminated stripe down
the frame as the rolling shutter camera captures the one or more
images, analyzes an illuminated section of the one or more images
to identify a region of interest in the illuminated section, and
stops sweeping of the illuminated stripe when the region of
interest is identified.
15. The biometric analysis system of claim 1 provided as a
smartphone having said one or more illumination sources, said
rolling shutter camera, and said adaptive trigger module.
16. A biometric analysis system, comprising: one or more
illumination sources configured to provide dim illumination to a
scene including a subject and configured to provide flash
illumination to the subject in the scene; a rolling shutter camera
configured to capture one or more images; and an adaptive trigger
module configured to (i) analyze the scene to detect eyes of the
subject in the scene during dim illumination of the scene, (ii)
identify the eyes of the subject as a region of interest, (iii)
determine a position in a frame of the rolling shutter camera that
coincides with the identified region of interest, and (iv) arrange
a flash pulse delay between a start of image writing by the rolling
shutter camera and a trigger of the one or more illumination
sources such that a stripe of the flash illumination coincides with
the identified region of interest.
17. The biometric analysis system of claim 16, wherein the flash
pulse delay ensures that the illuminated region of interest is
maintained in a center of the frame of the rolling shutter
camera.
18. The biometric analysis system of claim 16, comprising a
feedback module configured to analyze a captured image of the
region of interest and determine if the region of interest is
illuminated by the stripe of the flash illumination.
19. The biometric analysis system of claim 18, wherein the adaptive
trigger module is configured to adjust the flash pulse delay based
on the determination of the feedback module to ensure that the
region of interest is illuminated by the stripe of the flash
illumination.
20. The biometric analysis system of claim 16 provided as a
smartphone having said one or more illumination sources, said
rolling shutter camera, and said adaptive trigger module.
21-24. (canceled)
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] The present application claims the benefit of priority to
U.S. Provisional Patent Application No. 62/316,347, filed Mar. 31,
2016, which is hereby incorporated by reference in its
entirety.
TECHNICAL FIELD
[0002] The present disclosure relates to systems and methods of
biometric analysis and, in particular, to biometric analysis
systems including an adaptive trigger configured to align an object
(e.g., a subject, a barcode, or the like) within a camera
field-of-view to improve application of flash illumination to the
desired area of the object for capture and analysis.
BACKGROUND
[0003] Security is a concern in a variety of transactions involving
private information. As an example, iris recognition is a
well-accepted and accurate means of biometric identification used
in government and commercial systems around the world that enables
secure transactions and an added layer of security beyond keys
and/or passwords. Due to the increased security provided by iris
recognition systems, an increase in use of such systems has
occurred around the world.
[0004] Traditional cameras used in biometric identification are
generally expensive. Video cameras that use complementary
metal-oxide-semiconductor (CMOS) or semiconductor charge-coupled
device (CCD) image sensors typically use electronic shutters to
determine the time period over which the sensor measures light. A
trigger signal opens the shutter for a predetermined time during
which each pixel within the sensor array collects (integrates)
incoming light. At the end of the exposure, the signal collected
(integrated) in each pixel during the exposure remains fixed and is
then systematically read out, converted to a digital signal, and
processed to become an image. Pixels are then cleared and readied
for the next exposure to light.
[0005] There are different types of electronic shutters used in the
industry. FIG. 1 is a diagram of a traditional global shutter
camera schedule of events for signal collection, including the
timeline for exposure of each row of an image sensor in the global
shutter camera. Global shutters simultaneously expose an entire
array of pixels, e.g., N rows by M columns, during signal
collection. During a single exposure event (during t.sub.exp),
light is simultaneously collected in each pixel. When the global
shutter closes, the light signal within each pixel represents the
image during the period of the single exposure. All pixels
integrate signal over exactly the same period of time. Global
shutter cameras avoid flash timing issues incurred when using
rolling shutter cameras. However, global shutter cameras are
expensive options for biometric analysis, thereby increasing the
overall costs associated with biometric analysis systems.
[0006] Rolling shutter cameras save cost and size in their sensor
design. FIG. 2 is a diagram of a traditional rolling shutter
schedule of events for signal collection, including the rolling
shutter timeline. Rolling shutters expose an array of pixels
differently from global shutters. A rolling shutter system exposes
a first row of pixels for an exposure time (t.sub.exp) and then
commences to read-out the exposed row of pixels for digitization.
The read-out process occupies a unique onboard resource for a
period referred to as a read-out time during which no other row can
be read-out. To minimize the duration of the total exposure
including the read-out process, the rolling shutter exposes the
second row of pixels during a time that is equal to but delayed
from the first row by a read-out time. The second row is thereby
exposed to light and ready to be read-out at the moment that the
read-out process for the first row is complete. The third row is
exposed to light for a time interval equal in length to that of the
first two rows but delayed relative to the second row by a read-out
time allowing for the required time to read-out the second row. The
process "rolls" down the pixel array reading row-by-row in sequence
taking a total time equal to the exposure time for a single row
plus the read-out time interval, times the number of rows. The time
interval during which a row is exposed and therefore the events
captured by that row are different for each row for a rolling
shutter sensor. This is a key difference from a global shutter
sensor, especially when using a short flash.
[0007] As shown in FIGS. 1 and 2, the light collection time period
for each row of a sensor with a global shutter is simultaneous
while the time periods of light collection for each row of a sensor
equipped with a rolling shutter are not simultaneous. Rather, light
collection time periods for each row of a rolling shutter are
offset from one another with a delay between rows equal to the row
read-out time. The different exposure techniques result in image
shearing. For example, FIG. 3 shows an image in which a moving fan
blade was captured by an image sensor with a global shutter with no
or little distortion as compared to the same moving fan captured by
an image sensor with a rolling shutter shown in FIG. 4. Image
shearing is an inevitable consequence of the row-by-row time delays
built into a rolling shutter sensor in which each row "sees" the
scene over a slightly different and offset time interval.
[0008] The row-by-row delay in exposure of a rolling shutter also
has an effect on coordinating an exposure with flash illumination.
As discussed herein, a flash refers to a short, intense period of
illumination of a subject during which the light applied by the
flash dominates other sources of light on the scene (e.g., the area
surrounding the subject). FIG. 5 shows the exposure of each row
during t.sub.exp with the shaded region indicating the duration of
time of the flash illumination occurring simultaneous during
t.sub.exp. In a global shutter, the exposure of all the pixels in a
sensor can be coordinated with the application of the flash
illumination to the scene. For example, if the period of the flash
pulse is 1 ms, the global shutter can open simultaneously with the
start of the pulse and close simultaneously with the end of the
pulse 1 ms later. The flash illuminates the pixels during and only
during their global exposure. Light forming the image is, by
assumption, dominated by the light applied to the scene by the
flash. If, for example, sunlight is present in the scene, the
irradiance on the object from the flash is significantly brighter
than that of the sunlight during the flash pulse when pixels are
exposed by the global shutter.
[0009] Coordinating a flash pulse with a rolling shutter exposure
is more complicated than with a global shutter. FIG. 6 shows the
rolling shutter exposure during t.sub.exp for each row extending
diagonally across the diagram, and the time period for flash
illumination illustrated as the vertical shaded region t.sub.p.
After a flash delay, the flash pulse occurs between the start pulse
and end pulse points of the diagram. Because the exposure period
for each row is delayed from the previous row by a short read-out
time interval, illumination of a full frame requires that a flash
pulse provide illumination during a period when all rows are
integrating light. Failure to meet this condition creates a
situation in which some rows of pixels integrate light from the
flash pulse while some do not, and perhaps some rows integrate
light from only a portion of the flash pulse. In this case, the
image is unevenly illuminated. As shown in FIG. 6, some rows are
finished integrating before the flash starts and other rows do not
start integrating until after the flash ends. In addition, other
rows integrate a partial flash and some integrate the full flash.
Thus, a subset of lines on the rolling shutter sensor receive
adequate illumination, but outside of this set of lines, the other
parts of the sensor remain largely dark.
[0010] One example of a short flash pulse can be considered with
respect to FIG. 6, which, across the top horizontal line, four
horizontal dashed lines, and bottom lines, respectively shows row
numbers 0, 200, 450, 600, 850, and 1000. The flash pulse can start
as row 200 of 1000 rows finishes integrating the signal and begins
to read-out, indicated by the top dashed line. The flash pulse can
end as row 850 of 1000 begins integrating the signal, indicated by
the bottom dashed line. FIG. 6 shows that rows 450 through 599
receive the full illumination of the flash pulse, as bracketed by
the middle two dashed lines. However, rows 200 to 449 and rows 600
to 849 only receive a portion of the flash illumination while rows
outside of these ranges, e.g., rows 1 to 199 and 850 to 1000,
receive no flash illumination. Assuming insignificant ambient
light, the resulting image would show an illumination stripe
surrounded by dim regions. The transition from bright to dim at the
top and bottom of the stripe is due to rows that receive flash
illumination over a fraction of the total pulse time. FIG. 7 shown
a portion of an image acquired using a rolling shutter camera with
a delayed flash pulse in which the recorded irradiance is plotted
to show dark regions before and after the flash, ramp-up and
ramp-down regions of partial illumination, and a plateau region of
complete flash illumination. The plateau region of FIG. 7 is not
flat because the flash itself was not uniform over the
field-of-view. As another example, an image captured using a
rolling shutter camera would include a horizontal stripe with a
vertical height proportional to the duration of the flash
illumination. In cases with bright ambient illumination, the
un-flashed portion of the image would appear, but might be
significantly dimmer if the flash illumination is brighter than the
ambient illumination.
[0011] When an image of a particular object is desired with a
rolling shutter camera, a trigger signal can be initiated by the
sensor controller to fire the flash at a preset time relative to
the start of image writing. For example, the flash can fire when
the first line is written and can remain on for 50 of 1000 lines.
The resultant image would be flash illuminated for the top 5% of
the image and would be dark elsewhere. The same flash can be
delayed until the 500.sup.th line of 1000 lines, resulting in an
image with a stripe of illuminated content approximately halfway
down the frame. With such an arrangement, the photographer would
need to align the subject within the camera field-of-view such that
the stripe of illumination detected by the sensor corresponds to
the position of the desired object.
[0012] Traditionally, one solution to the problem of flash
illuminating an image using a rolling shutter has been to use an
extended period of illumination, e.g., a flash pulse that is
started simultaneously with the beginning of the exposure of the
first row of pixels and is not finished until the last row of
pixels has been exposed. The extended period of illumination is
needed to expose the entire image since image lines are written
sequentially rather than all at once (as is the case with a camera
including a more expensive and physically larger global shutter
sensor). This technique necessitates a longer flash pulse compared
to the global shutter case, and would show up in FIG. 6 as a shaded
region covering all of the rows with a duration equal to the frame
time. This technique would also illuminate the full frame shown in
FIG. 7. Additional requirements on the flash in terms of power
output, heating and reliability are needed based on the longer
pulse for this technique. A longer pulse might also challenge
requirements for eye-safety. For these reasons, full frame pulses
with rolling shutters are considered impractical.
[0013] Thus, a need exists for improved biometric analysis systems
including a rolling shutter that are capable of illuminating and
capturing the desired area of an object for identification without
an extended flash. These and other needs are addressed by the
systems and methods of biometric analysis of the present
disclosure.
SUMMARY
[0014] In accordance with embodiments of the present disclosure, an
exemplary biometric analysis system is provided that includes one
or more illumination sources configured to provide dim illumination
to a scene including an object, and further configured to provide
flash illumination to the object in the scene. In some embodiments,
the dim illumination can be provided by an illumination source
external and separate from the biometric analysis system, such as
ambient light, sunlight, any other light source, or the like (e.g.,
one or more of the illumination sources can be ambient light). In
some embodiments, the biometric analysis system can include a
single illumination source that provides the dim illumination, with
the same illumination source providing the flash illumination at
the determined time period. In some embodiments, the biometric
analysis system can include a first illumination source that
provides the dim illumination, and a second (separate) illumination
source that provides the flash illumination. In some embodiments,
the first illumination source can continue to provide the dim
illumination during the flash illumination from the second
illumination source. In some embodiments, the first illumination
source can be automatically actuated into a non-illuminating
configuration during the flash illumination provided by the second
illumination source, and automatically actuated into an
illuminating configuration after the flash illumination is
complete.
[0015] The biometric analysis system includes a rolling shutter
camera configured to capture one or more images. The rolling
shutter camera generally includes a frame with a field-of-view. The
term "image" as used herein can include still frame images, video,
combinations thereof, or the like. The biometric analysis system
includes an adaptive trigger module configured to be executed by a
controller or processing device. The adaptive trigger module, when
executed, can be configured to analyze the scene to detect the
object in the scene during dim illumination of the scene. The
adaptive trigger module, when executed, can be configured to
determine a position in the frame of the rolling shutter camera
that coincides with the detected object in the scene. The adaptive
trigger module, when executed, can be configured to arrange a delay
(a time delay) between a start of image writing by the rolling
shutter camera and a trigger of the one or more illumination
sources such that a stripe of the flash illumination coincides with
the detected object in the scene.
[0016] In some embodiments, the adaptive trigger module, when
executed, can be configured to track movement of the object within
a field-of-view of the rolling shutter camera. In such embodiments,
the adaptive trigger module can be configured to modify the delay
between the start of image writing by the rolling shutter camera
and the trigger of the one or more illumination sources based on
detected movement of the object within the field-of-view.
[0017] In some embodiments, the adaptive trigger module can be
configured to detect a region of interest of the object and
arranges the delay such that the stripe of flash illumination
coincides with the detected region of interest of the object. In
one embodiment, the region of interest of the object can include
one or both eyes of a person.
[0018] In some embodiments, the object can be a person. In such
embodiments, the adaptive trigger module includes a face finder
configured to detect a face (and/or features of the face) of the
person. In some embodiments, the object can be a physical item. In
such embodiments, the adaptive trigger module can include an
identifier finder configured to detect a unique identifier (e.g., a
barcode, a quick response (QR) code, combinations thereof, or the
like) associated with the physical item.
[0019] In some embodiments, the one or more illumination sources
can be configured to provide the flash illumination as a
synchronized pulse of flash illumination. The flash illumination
provided by the one or more illumination sources is brighter than
the dim illumination provided by the one or more illumination
sources. In one embodiment, the one or more illumination sources
can be near infrared (NIR) illumination sources. In some
embodiments, the adaptive trigger module can be configured to sweep
an illuminated stripe down the frame as the rolling shutter camera
captures the one or more images, analyze an illuminated section of
the one or more images to identify a region of interest in the
illuminated section, and stop sweeping of the illuminated stripe
when the region of interest is identified.
[0020] In accordance with embodiments of the present disclosure, an
exemplary biometric analysis system is provided that includes one
or more illumination sources configured to provide dim illumination
to a scene including a subject and configured to provide flash
illumination to the subject in the scene. The biometric analysis
system includes a rolling shutter camera configured to capture one
or more images. The biometric analysis system includes an adaptive
trigger module configured to be executed by a controller or
processing device. The adaptive trigger module, when executed, can
be configured to analyze the scene to detect eyes of the subject in
the scene during dim illumination of the scene. The adaptive
trigger module, when executed, can be configured to identify the
eyes of the subject as a region of interest. The adaptive trigger
module, when executed, can be configured to determine a position in
a frame of the rolling shutter camera that coincides with the
identified region of interest. The adaptive trigger module, when
executed, can be configured to arrange a flash pulse delay between
a start of image writing by the rolling shutter camera and a
trigger of the one or more illumination sources such that a stripe
of the flash illumination coincides with the identified region of
interest. In some embodiments, rather than or in addition to using
dim illumination, the adaptive trigger module can be executed to
search and find the region of interest by systematically sweeping
an illuminated stripe down the field-of-view and stopped when a
region of interest is detected. Thus, rather than using a single
frame (as in dim illumination), the illuminated stripe can be a
fraction in width of the full frame and is advanced across the
field-of-view, thereby using multiple frames to detect the region
of interest. The region of interest can be tracked by the adaptive
trigger module after being detected.
[0021] The flash pulse delay can ensure that the illuminated region
of interest is substantially maintained in a center of the frame of
the rolling shutter camera. The biometric analysis system can
include a feedback module configured to be executed by the
controller or processing device. The feedback module, when
executed, can be configured to analyze a captured image of the
region of interest and determine if the region of interest is
illuminated by the stripe of the flash illumination. The adaptive
trigger module, when executed, can be configured to adjust the
flash pulse delay based on the determination of the feedback module
to ensure that the region of interest is illuminated by the stripe
of the flash illumination.
[0022] In accordance with embodiments of the present disclosure, an
exemplary method of biometric analysis is provided. The method
includes illuminating a scene and an object in the scene with dim
illumination from one or more illumination sources. The method
includes analyzing the scene with an adaptive trigger module to
detect the object in scene during dim illumination. The method
includes determining a position in a frame of a rolling shutter
camera that coincides with the detected object in the scene. The
method includes arranging a delay between a start of image writing
by the rolling shutter camera and a trigger of the one or more
illumination sources such that a stripe of flash illumination
provided by the one or more illumination sources coincides with the
detected object in the scene.
[0023] In some embodiments, the method can include tracking
movement of the object within a field-of-view of the rolling
shutter camera with the adaptive trigger module. In such
embodiments, the method can include modifying the delay between the
start of image writing by the rolling shutter camera and the
trigger of the one or more illumination sources with the adaptive
trigger module based on detected movement of the object within the
field-of-view.
[0024] In accordance with embodiments of the present disclosure, an
exemplary non-transitory computer-readable medium storing
instructions for biometric analysis is provided. Execution of the
instructions by a processing device (or controller) causes the
processing device to illuminate a scene and an object in the scene
with dim illumination from one or more illumination sources.
Execution of the instructions by a processing device causes the
processing device to analyze the scene with an adaptive trigger
module to detect the object in scene during dim illumination.
Execution of the instructions by a processing device causes the
processing device to determine a position in a frame of a rolling
shutter camera that coincides with the detected object in the
scene. Execution of the instructions by a processing device causes
the processing device to arrange a delay between a start of image
writing by the rolling shutter camera and a trigger of the one or
more illumination sources such that a stripe of flash illumination
provided by the one or more illumination sources coincides with the
detected object in the scene.
[0025] Other objects and features will become apparent from the
following detailed description considered in conjunction with the
accompanying drawings. It is to be understood, however, that the
drawings are designed as an illustration only and not as a
definition of the limits of the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
[0026] To assist those of skill in the art in making and using the
disclosed systems and methods of biometric analysis, reference is
made to the accompanying figures, wherein:
[0027] FIG. 1 (prior art) is a diagram of a traditional global
shutter camera schedule of events showing simultaneous exposure of
each row between start and end of exposure followed by simultaneous
read-out.
[0028] FIG. 2 (prior art) is a diagram of a traditional rolling
shutter camera schedule of events showing start of frame, cascading
exposure of pixel rows, and rolling read-out, with delay between
rows being equal to read-out time.
[0029] FIG. 3 (prior art) is an image of a moving fan illustrating
minimal distortion as captured by a traditional global shutter
camera.
[0030] FIG. 4 (prior art) is an image of a moving fan illustrating
shearing distortion effects caused by cascading row delays as
captured by a traditional rolling shutter camera.
[0031] FIG. 5 (prior art) is a diagram of a traditional global
shutter schedule of events showing simultaneous exposure of each
row between start and end of exposure followed by simultaneous
read-out, with a flash firing during the same time interval as that
in which the rows are exposed (t.sub.exp).
[0032] FIG. 6 (prior art) is a diagram of a traditional rolling
shutter schedule of events showing start of frame, cascading
exposure of pixel rows, and rolling read-out, with a flash firing
after the beginning of the frame, illuminating some but not all
rows, and some rows partially but others completely, resulting in a
ramp-up, plateau and ramp-down periods of flash illumination.
[0033] FIG. 7 (prior art) is a diagram of a vertical strip of an
image (re-oriented horizontally to align with the time axis)
showing dark, ramp-up, plateau, ramp-down, and another dark region
due to timing of flash in a traditional rolling shutter frame.
[0034] FIG. 8 is a schematic block diagram of an exemplary
biometric analysis system in accordance with the present
disclosure.
[0035] FIG. 9 is a representation of an image captured by a rolling
shutter camera of the exemplary biometric analysis system including
a stripe of flash illumination across a region of interest.
[0036] FIG. 10 is a diagrammatic view of an exemplary biometric
analysis system in accordance with the present disclosure.
[0037] FIG. 11 is a diagrammatic view of a frame action of a
rolling shutter camera of an exemplary biometric analysis system in
accordance with the present disclosure.
[0038] FIG. 12 is a flowchart illustrating an exemplary process of
implementing an exemplary biometric analysis system in accordance
with the present disclosure.
[0039] FIG. 13 is a flowchart illustrating an exemplary process of
implementing an exemplary biometric analysis system in accordance
with the present disclosure.
[0040] FIG. 14 is a block diagram of an exemplary computing device
for implementing an exemplary biometric analysis system in
accordance with the present disclosure.
[0041] FIG. 15 is a block diagram of an exemplary biometric
analysis system environment in accordance with the present
disclosure.
DESCRIPTION OF EXEMPLARY EMBODIMENTS
[0042] In accordance with embodiments of the present disclosure,
exemplary biometric analysis systems are described herein for a
rolling shutter camera that include an adaptive trigger for
ensuring that the stripe of light from the flash illumination
illuminates a region of interest of an object, such as the eyes of
a person, during image writing. The biometric analysis systems
generally include one or more illumination sources that initially
provide dim illumination to a scene to detect the object and
identify the region of interest of the object. Based on the
identified region of interest, a delay between the start of image
writing by the rolling shutter camera and the trigger of the
illumination source is arranged or coordinated to ensure that the
stripe of the flash illumination coincides with the region of
interest of the object.
[0043] Cameras that include rolling shutter image sensors are
typically cheaper than those with global shutter image sensors. The
exemplary biometric analysis systems use flash illumination in
combination with inexpensive rolling shutter cameras of the type
found in low-cost cameras and smartphones. The biometric analysis
systems address a rolling shutter-specific problem that occurs when
the brightness of the ambient light in the scene may need an
exposure time less than the full frame time to avoid saturation,
and the exposure time is sufficiently short that the first rows of
the image are finished integrating light and reading out before the
last lines of the image have begun integration. In cases of short
exposure times, a single, short flash pulse will not illuminate an
entire frame, leaving portions of the frame relatively dark. Long
exposure times can result in motion blur.
[0044] The exemplary biometric analysis systems balance the
following parameters to ensure the captured image quality is
sufficient for identification, and further ensuring that the region
of interest is adequately illuminated by the flash: sensor frame
rate, sensor exposure time, flash pulse intensity, flash pulse
duration, flash pulse repetition, processed frame rate, and flash
pulse offset. Increasing the sensor frame rate over the available
ranges increases the speed that the reset and read lines move over
the image. For a given sensor exposure time, increasing the frame
rate effectively increases the number of exposed lines during that
period. For example, with the exposure time remaining the same, the
execution time of the internal clock of the camera can be increased
to operate the camera at a faster rate, thereby capturing a greater
number of lines during the exposure time. The drawbacks of
increased sensor frame rates are additional data transfers and
additional heating due to higher clock speed. Lower frame rates may
be desirable for dark indoor lighting conditions and higher frame
rates may be desirable for outdoor bright lighting conditions.
[0045] Increasing the sensor exposure time increases the distance
between the reset and read times. This effectively increases the
number of exposed lines during that period, allowing a short flash
pulse to cover more lines. The drawback of increasing this period
is that additional ambient light is accumulated over the entire
period. During outdoor operation, it may be beneficial to keep the
exposure time as short as possible, but allowing enough lines to be
covered by the flash. Keeping the exposure time shorter is also
important to avoid introducing motion blur which can degrade
recognition performance.
[0046] The flash pulse needs to be bright enough to overpower
direct sunlight for effective outdoor operation. The flash pulse is
limited by available system power and eye safety limits. In
general, eye safety can be roughly proportional to the total amount
of energy absorbed by the eye over a period of time (e.g.,
t.sup.-0.75). Reducing the intensity, pulse duration, and pulse
repetition all move the system into a safer standing. Longer flash
pulses allow more lines to be exposed in a rolling shutter.
However, longer pulses can lead to increased eye safety concerns
forcing the system to use less intense pulses when longer durations
are used.
[0047] Several short flash pulses may be used back-to-back to
illuminate additional lines with minimal overlap between exposed
lines. This technique can reduce the total eye exposure and power
requirements while expanding the total number of lines that can be
properly illuminated. Pulse rate, duration, and intensity can all
impact the system power requirements. Increasing any of these
parameters generally increases the demand for power.
[0048] In some embodiments, since eye safety depends on the total
exposure over a period of time, one method to reduce the total
exposure can be to selectively process frames. While images can be
captured at 60 frames per second (fps), only one in four can be
illuminated with the flash, which would reduce the exposure to an
equivalent of 15 fps with the advantages of capturing at 60 fps.
Such technique allows for improved line read speeds, bringing the
rolling shutter camera closer to the global shutter camera
timing.
[0049] In some embodiments, the systems can include a continuous
feedback loop to a user who is attempting to align their eye with a
small box on the screen (e.g., in situations involving a smaller
screen, such as a smartphone). The feedback loop allows the system
to predetermine which lines should be exposed. A low level of
continuous, dim illumination can be provided to ensure the user has
a reasonable image to complete this task. Depending on the
available processing power and flash offset controls, the
continuously illuminated frames can be processed to detect specular
reflections, and further used to change the flash offset or delay.
Such operation allows the user additional margin when attempting to
place their eye in the box while keeping the exposure levels to a
minimum.
[0050] The biometric analysis systems adjust the portion of the
frame that is illuminated such that the important portion of the
image (e.g., a region of interest) is well-lit and the unimportant
portions of the images are left dim. The biometric analysis systems
can be used in imaging in the near infrared (NIR) spectrum using a
short, intense, eye-safe NIR flash with an NIR filter to freeze
motion and emphasize the flash relative to ambient light (including
bright sun). The biometric analysis systems including the NIR flash
with a rolling shutter camera can be used in, e.g., iris
biometrics, or the like, in which cost reduction realized by using
a rolling shutter instead of a global shutter is a primary driver
in component selection.
[0051] The biometric analysis systems include an adaptive step in
which dim illumination allows a vision system (e.g., a sensor in a
rolling shutter camera including a processor) to detect the
position of the desired object in the field-of-view of the camera.
In some embodiments, the dim illumination remains on for the entire
process of capturing an image, and reveals an under-illuminated
frame. Dim illumination, although dimmer than illumination by the
flash, is bright enough to perform an automated scene analysis,
e.g., face finding, object identification, or the like, that can
output a recommendation of a position in the frame that would
benefit from flash illumination. For example, if a face appeared
between 20% and 40% of the frame height measured from the bottom, a
controller or processor can arrange a delay between the start of
image writing and the flash trigger such that a stripe of
illumination coincides with the desired object. If the object moves
in the field-of-view, biometric analysis systems can function
dynamically to track movement of the object and change the delay
accordingly. In some embodiments, rather than or in addition to
using dim illumination, the adaptive trigger module can be executed
to search and find the region of interest by systematically
sweeping an illuminated stripe down the field-of-view and stopped
when a region of interest is detected. Thus, rather than using a
single frame (as in dim illumination), the illuminated stripe can
be a fraction in width of the full frame and is advanced across the
field-of-view, thereby using multiple frames to detect the region
of interest. The region of interest can be tracked by the adaptive
trigger module after being detected. Such arrangements to locate
and track the object and indicate to the adaptive trigger module
changes in the position of the object permits moving objects to be
accurately imaged by the biometric analysis systems.
[0052] The biometric analysis systems disclosed herein can be used
for rolling shutter photography (e.g., NIR photography) of objects
while outdoors or indoors. The biometric analysis systems can
acquire NIR images with controlled illumination in full sun or in
dim ambient lighting. An image of a desired object that fills less
than a full frame, e.g., a person's face with the entire person in
the field-of-view, can be captured by the biometric analysis
systems. Due to the low cost, small size and availability of
rolling shutter sensors, incorporation of the biometric analysis
systems into cost-sensitive or size-sensitive platforms can benefit
cost-sensitive or size-sensitive users. As one example, the
biometric analysis systems can be used in cargo identification in
which infrared dye for markings signify authenticity. As another
example, a box marked with writing can be imaged with a small and
inexpensive handheld infrared camera that would look for the
marking surrounding an NIR watermark.
[0053] The flash trigger can be adjusted to fire in order to
illuminate the markings, leaving the rest of the field-of-view
above and below the mark of interest in the dark. Handheld devices,
including smartphones, barcode readers, and other mass produced
electronic devices, can incorporate the biometric analysis systems
at a small and low cost. Traditionally, a rolling shutter camera
would reduce the cost for such systems, but would require an
extended flash to illuminate an entire scene to guarantee capturing
the subject of interest. The exemplary biometric analysis systems
include a rolling shutter camera with an adaptive image
content-based trigger that issues a shorter flash of a near
infrared light that allows for use of smaller batteries, provides
for a longer battery life, provides for improved eye safety, and
allows for longer illumination source lifetimes.
[0054] The biometric analysis systems disclosed herein can provide
flash illumination for illuminating an object sufficiently to
outshine the sun, and in some embodiments can use strobing flash
illumination as is disclosed in, e.g., U.S. Pat. Nos. 7,542,628;
7,627,147; 7,657,127; 9,131,141; U.S. Patent Application
Publication No. 2015/0098630; U.S. Patent Application Publication
No. 2016/0014121; U.S. Patent Application Publication No.
2016/0012218; and U.S. Patent Application Publication No.
2016/0012292, which are incorporated by reference herein.
[0055] The exemplary biometric analysis systems use a rolling
shutter camera and a flash illumination that is triggered
adaptively to illuminate the region of interest of the scene or
object. The adaptive trigger can be configured to follow or track
the subject, and adjusts the flash timing to maintain the stripe of
flash illumination on the subject of interest. In one embodiment, a
face finder can be used as an image analytic to trigger the flash.
In one embodiment, a unique identifier, such as a bar code or QR
code, can be detected and illuminated by the stripe of flash
illumination. In either case, the biometric analysis system can
locate the object in a relatively dimly lit scene, and then
initiates the repetitive flash that illuminates the object. Flash
timing can be adjusted to maintain the object in the illuminated
region of the field-of-view of the rolling shutter camera.
[0056] With respect to the delayed flash, an assumption can be made
that the start of integration of the first row of a rolling shutter
sensor coincides with time t=0. A further assumption can be made
that a flash pulse is started at some time later, e.g., delayed by
a time .DELTA.t. Rows that are read-out (e.g., integrate light)
before At miss any light from the flash. Rows that are in the
process of integrating light when the flash starts--some nearly
finished, some just starting--receive fractions of the flash and
form the ramp-up region of partial illumination. Pixels in rows
that start integrating at or after the start of the flash and that
finish integrating light by the end of the flash are illuminated by
the full flash. By adjusting the delay time At, the rows of full
flash illumination can be adjusted from the top of the image frame
to the bottom to focus on a region of interest located within the
frame.
[0057] In particular, the biometric analysis systems can be used to
detect the region of interest within the frame of the rolling
shutter camera, and intentionally delay the flash illumination in a
coordinated manner to ensure that the region of interest is
illuminated by the flash during exposure and read-out of the
corresponding rows. Proper adjustment of the delay time can cause
the stripe of flash illumination due to the short flash pulse to
brighten the region of interest. In some embodiments, the stripe
can scan or search the image in successive frames of a photo or
video stream. Such technique can be performed by systematically
sweeping an illuminated stripe down the field-of-view and stopping
when a region of interest is detected. Thus, rather than using a
single frame (as in dim illumination), the illuminated stripe can
be a fraction in width of the full frame and is advanced across the
field-of-view, thereby using multiple frames to detect the region
of interest. The region of interest can be tracked by the adaptive
trigger module after being detected.
[0058] In some embodiments, a pre-flash (e.g., a single long, low
intensity flash that illuminates the full frame and that is both
eye-safe and within the operating range of the illumination
hardware) can provide a single, dim image of the scene. Automated
image analysis of the full scene that can tolerate the low light
can detect and output a region of interest that is the width of the
image and the height of the flash illumination stripe. With the
region of interest defined, an automated delay signal can set the
illumination stripe to the optimal position to illuminate the
region of interest in succeeding frames of a photo or video stream
using the full flash brightness over the narrowed region of
interest. If the desired feature of the object begins to move out
of the illumination stripe (as determined by video analysis
performed at the frame rate), the delay can adapt using negative
feedback to maintain the position of the desired feature centered
within the well-lit region of the image.
[0059] In some embodiments, the biometric analysis systems can be
used to capture NIR images of an object that is bathed in sunlight.
While a fraction of the solar spectrum covers the NIR wavelengths,
it can be assumed that filtered solar NIR irradiance (e.g., from
800 nm to 900 nm) is insufficient to illuminate a video frame that
is adequately short in duration to suppress motion blur. It can be
further assumed that using a long flash that covers the entire
rolling shutter frame period is impractical, even with a short
exposure time, due to hardware limitations or eye safety concerns.
A short NIR flash pulse that can only illuminate a portion of the
frame with an adaptive delay time to ensure that the region of
interest is illuminated can therefore be used.
[0060] The biometric analysis systems initially detect and identify
a region of interest of an object, such as the eyes of a subject.
Eye finding can be accomplished using a number of different
processes, such as full frame NIR illumination with a relatively
dim, eye-safe (and hardware-safe) long NIR flash pulse; partial
frame illumination due to a short, intense pulse that is
automatically swept frame-by-frame over the full field-of-view
until the eyes appear in the search; and/or image acquisition by a
color (or other spectrum) auxiliary camera that produces a full
frame image allowing eye-finding or face-finding, and then
translation from the auxiliary camera frame position to a NIR
camera position.
[0061] After the eyes are detected and their position located in
the NIR camera frame, the biometric analysis systems automatically
set the flash pulse delay to illuminate a stripe across the frame
including the eyes (e.g., the region of interest). The delay that
sets the vertical position of the illumination stripe can be
automatically adjusted to maintain the eyes in substantially the
center of the frame. While the portions of the frame outside of the
illumination stripe remain unusably dark, the important information
about the eyes remains well lit.
[0062] The competing solar illumination may dimly illuminate the
rest of the frame. This is not an issue as long as the flash
illumination stripe remains resolved and does not saturate due to
the sum of the in-band solar irradiance and the applied NIR light.
If the illumination stripe nears the saturation level, the
intensity of the applied flash can be throttled back. It is
preferable to decrease the NIR illumination irradiance on the
object rather than reducing the exposure or pulse duration, since
the later strategies reduce the portions of the image that are
illuminated.
[0063] Any spectrum of light can be used as long as the camera uses
a rolling shutter with an in-band flash illuminator. For example,
visible light, NIR, short-wave infrared (SWIR), long-wave infrared
(LWIR), ultraviolet (UV), or multi-spectral cameras with rolling
shutters that use flash illumination. Any process of finding the
region of interest can be used as long as the process returns a
coordinate to the biometric analysis system that indicates the time
delay between the start of frame and the start of the flash. For
example, use of an eye finder that seeks corneal reflections of the
flash could be useful if the region of interest contains one or two
eyes of the subject. As a further example, a color face camera
could be useful to find the eyes of a subject in iris recognition;
but if nose-recognition were used, the region of interest would
contain the subject's nose. If the same camera were used to find
the region of interest and to acquire the information from the
region of interest, a different illumination protocol could be used
for each function. For example, a long, relatively dim pulse could
illuminate the entire frame, followed by a short, bright pulse to
illuminate the region of interest with better fidelity (e.g.,
improved signal-to-noise ratio). Any process that provides the
information needed to set the delay time would also be useful.
[0064] In some embodiments, feedback can be used to maintain the
region of interest in the illumination stripe. For example,
negative feedback can be used to correct the position of the region
of interest within the illumination stripe. If the object of
interest begins to move or slip to the top or bottom of the region
of interest illuminated by the flash, feedback can be used to
correct the delay time. For example, if the stripe illuminating the
region of interest needs to move up in the frame to follow the
object of interest, the delay would need to be corrected in one
direction. If the object of interest moves down in the frame, the
sign of the delay correction would need to be in the opposite
direction. If the delay reaches the minimum or maximum value or, in
other words, the object of interest begins to leave the camera
field-of-view, the controller or processing device automatically
reorients the camera to maintain the object in the field-of-view.
In some embodiments, a feedback signal (e.g., an audio alarm, a
visual alarm, combinations thereof, or the like) can provide
feedback to an operator regarding the object leaving the
field-of-view.
[0065] The width of the flash illumination stripe relative to the
frame height can be set by practical considerations. For example,
rolling shutter cameras with very short read-out times generally
have wider flash illumination stripes, all other things being
equal. The longest flash pulse can be dictated by practical issues,
including available power to the illumination source, operating
conditions in power, and temperature of the illumination source.
For example, it is known that flashing an illumination source too
often can damage it, and reduces the safety of human eyes in and
around the illumination source. A flash illumination stripe that is
narrow relative to the frame height generally should be more
accurately timed than a flash illumination stripe that is wider and
that can tolerate more motion of the object of interest within the
flash illumination stripe.
[0066] With reference to FIG. 8, a schematic block diagram of an
exemplary biometric analysis system 100 (hereinafter "system 100")
is provided. The system 100 is configured to detect and identify a
region of interest of an object, such as the eyes of a person, and
generates a time delay of the flash illumination such that the
flash coincides with the region of interest. A rolling shutter
camera can therefore be used for capturing one or more images for
biometric analysis with the delayed flash illumination ensuring
that the region of interest will be properly illuminated in the
captured images. The system 100 is configured to track the region
of interest within the frame or field-of-view of the rolling
shutter camera, and adjusts the time delay of the flash pulse
appropriately based on movement of the region of interest in and
out of the frame.
[0067] The system 100 generally includes one or more illumination
sources 102. The illumination sources 102 are configured to
initially provide dim illumination 104 to a scene including an
object to detect and identify the object (or the region of interest
of the object). In some embodiments, the dim illumination 104 can
be separate from the system 100 and can be provided by, e.g.,
ambient light, any other illumination source, sunlight,
combinations thereof, or the like. The illumination sources 102 are
further configured to provide flash illumination 106 (e.g., a pulse
of flash illumination) to illuminate the region of interest during
exposure of the rows corresponding to the region of interest,
thereby ensuring that the region of interest will be illuminated in
the captured image. In some embodiments, a single illumination
source 102 can provide both the dim illumination 104 and the flash
illumination 106. In some embodiments, separate illumination
sources 102 can provide the dim illumination 104 and the flash
illumination 106. In one embodiment, the illumination sources 102
can be NIR illumination sources. It should be understood that the
dim illumination 104 is less bright than the flash illumination
106, and provides sufficient ambient light for the system 100 to
detect objects in a scene as viewed by the rolling shutter
camera.
[0068] The system 100 includes a rolling shutter camera 108
including a frame 110 with a field-of-view. The rolling shutter
camera 108 is configured to capture one or more images (e.g., still
frame images, video images, combinations thereof, or the like). The
rolling shutter camera 108 is configured to initially capture an
image of the scene and the object in the scene under dim
illumination 104 conditions. The system 100 includes an adaptive
trigger module 112 configured to be executed by a processing device
114 having a processor 116 (e.g., a controller). Although
illustrated as separate components from the rolling shutter camera
108, it should be understood that in some embodiments, the rolling
shutter camera 108 can include the illumination sources 102, the
adaptive trigger module 112, the processing device 114, or the
like. In some embodiments, the processing device 114 can be part of
a central computing system 118. When executed, the adaptive trigger
module 112 can analyze the captured image of the scene illuminated
by the dim illumination 104 and detects the region of interest 120
within the scene. For example, the adaptive trigger module 112 can
detect and identify the region of interest 120 as one of more eyes
of the user, and stores the rows of the frame 110 corresponding to
the region of interest 120 in a database 122 (e.g., a local or
remote database).
[0069] Based on the rows of the frame 110 corresponding to the
region of interest 120, the adaptive trigger module 112 arranges a
flash pulse delay 124 between the start of image writing by the
rolling shutter camera 108 and a trigger of the flash illumination
106 such that the flash illumination 108 coincides with the
identified region of interest 120. For example, if the eyes of the
user are the region of interest 120 corresponding with rows 400-600
of the frame 110, the adaptive trigger module 112 arranges the
flash pulse delay 124 such that the flash illumination 106 is
triggered to illuminate the object during rows 400-600. Thus,
rather than triggering the flash illumination 106 at the same time
as the start of image writing by the rolling shutter camera 108,
the image writing can start before triggering of the flash
illumination 106 and the flash illumination 106 is only triggered
to illuminate the region of interest 120. Such flash delay 124
ensures that the region of interest 120 is illuminated in the
images 126 captured by the rolling shutter camera 108. The
resulting image 126 generally includes a stripe of flash
illumination 106 extending across the image 126 and over the eyes
of the user (see, e.g., FIG. 9). Depending on the length of the
exposure time, the height of the stripe of flash illumination 106
can be greater or smaller relative to the top and bottom of the
image 126.
[0070] In some embodiments, the adaptive trigger module 112 can
include a face finder algorithm that locates the face and/or
features of the face in the scene. The face finder algorithm can
therefore locate the face identifies one or more eyes of the face
as the region of interest 120. Although discussed herein as eyes of
a user, it should be understood that the region of interest 120 can
be a unique identifier associated with a physical item that can be
detected by an identifier finder algorithm of the adaptive trigger
module 112. For example, the identifier finder algorithm of the
adaptive trigger module 112 can be used to detect, e.g., a barcode,
a QR code, or the like, on a physical item, identifies such unique
identifier as the region of interest 120, and adjusts the flash
delay 124 to ensure that the unique identifier is illuminated
during capture of the image 126.
[0071] The adaptive trigger module 112 can be configured to track
movement of the object within the field-of-view or frame 110 of the
rolling shutter camera 108. In some embodiments, the adaptive
trigger module 112 can be configured to continuously scan the scene
in the frame 110 to detect the position of the region of interest
120. If the user moves such that the region of interest 120 in the
frame 110 is not centered or changes relative to the previously
detected rows for the region of interest 120, the adaptive trigger
module 112 can determine the new rows corresponding to the region
of interest 120, stores the new rows in the database 122, and
adjusts the flash delay 124 to coincide with the new rows for the
region of interest 120. Thus, an object moving in the frame 110 of
the rolling shutter camera 108 can be tracked by the adaptive
trigger module 112 to allow for the region of interest 120 of a
moving object to be properly illuminated by the flash illumination
106.
[0072] The system 100 includes a feedback module 128 configured to
be executed by the processing device 114 and/or the central
computing system 118. In some embodiments, the rolling shutter
camera 108 can include the feedback module 128. When executed, the
feedback module 128 receives as input the images 126 captured by
the rolling shutter camera 108 and analyzes the region of interest
120 captured in the image 126 to determine if the region of
interest 120 is illuminated by the stripe of flash illumination
106. If the region of interest 120 is illuminated by the stripe of
flash illumination 106, the feedback module 128 electronically
outputs such findings to the adaptive trigger module 112 and no
changes are made in the flash delay 124. If the region of interest
120 is not illuminated (or only partially illuminated) by the
stripe of flash illumination 106, the feedback module 128
electronically outputs such findings to the adaptive trigger module
112. Based on the feedback from the feedback module 128, the
adaptive trigger module 112 analyzes the frame 110 to locate the
region of interest 120, determines the new rows associated with the
region of interest 120, and adjusts the flash pulse delay 124 to
trigger the flash illumination 106 at the appropriate rows
coinciding with the region of interest 120. The feedback loop can
be continuously performed to ensure the region of interest 120 is
always illuminated by the stripe of flash illumination 106.
[0073] The system 100 can include a user interface 130 with a
graphical user interface (GUI) 132 for receiving input from a user
and for outputting information to the user. For example, the user
interface 130 can be used to initiate the process performed by the
system 100. As a further example, the user interface 130 can output
an audio and/or visual alarm to the user when the region of
interest 120 is no longer centered in the frame 110. The system 100
can include a communication interface 134 configured to provide a
means for electronic transmission between components of the system
100, e.g., the illumination sources 102, the processing device 114,
the rolling shutter camera 108, the adaptive trigger module 112,
the database 122, the user interface 130, the central computing
system 118, and feedback module 128, or the like.
[0074] FIG. 10 is a diagrammatic view of an exemplary biometric
analysis system 200 (hereinafter "system 200") of the present
disclosure. The system 200 includes an illumination source 202 that
provides dim illumination to the subject 204 in the scene. In some
embodiments, the illumination source 202 can provide 1 ms NIR
pulses to the scene. In some embodiments, the illumination source
202 can be an NIR light source not connected to the camera 208. The
illumination source 202 can be positioned a distance 206 from the
subject 204. In some embodiments, the illumination source 202 can
also be configured to provide the flash illumination to the subject
204.
[0075] The system 200 includes a camera 208 including a rolling
shutter (e.g., a rolling shutter camera, such as a smartphone
camera, or the like) configured to capture one or more images of
the subject. The camera 208 can have a field-of-view represented by
area 216, including the region of interest of the subject 204. In
some embodiments, the camera 208 includes an illumination source
210 (e.g., an integrated NIR light source) configured to provide
flash illumination to a region of interest (e.g., the eyes of the
subject 204). As an example, area 212 indicates the coverage of the
NIR flash illumination provided by the illumination source 210,
with the region of interest on the subject 204 being within the
area 212. The camera 208 can be spaced by a distance 214 from the
subject 204. In FIG. 10, the illumination source 202 appears
separate from camera 202, however, one skilled in the art shall
appreciate that, in some embodiments, the illumination source 202
can be incorporated into the camera 208 assembly such that the
camera 208 assembly is capable of providing both the dim
illumination and flash illumination to the subject 204.
[0076] FIG. 11 is a diagrammatic view of a frame action 300 of a
rolling shutter camera of the exemplary biometric analysis systems
disclosed herein. The frame action 300 includes a total frame time
302. Ambient light 304 having a low illumination level 306 can be
provided for all or part of the total frame time 302 by a dim
illumination source and/or sunlight. The total camera exposure time
308 indicates when the rolling shutter camera begins and ends image
writing and read-out.
[0077] Based on the determination of the region of interest and the
corresponding rows of the frame, a pulse of flash illumination 310
(e.g., NIR from a light-emitting diode (LED) illumination source)
is triggered to coincide with the rows of the region of interest.
The pulse of flash illumination 310 has a time or pulse width 312
that is significantly smaller than the camera exposure time 308,
and an illumination level 314 that is brighter or higher than the
level 306 of the ambient light 304. The region of interest is
thereby illuminated by the flash illumination 310 while the
remaining rows of the captured image are only illuminated by the
ambient light 304.
[0078] In some embodiments, a 1280.times.960 sensor can be used
with the rolling shutter camera. In some embodiments, a 120 fps VGA
(640.times.480) sensor can be used with the rolling shutter camera
to allow for full frame exposure in 4 ms, and can be synchronized
with an approximately 4 ms LED flash every 8.sup.th frame. The VGA
output can provide (unscaled) 15 fps feedback to the user facing
the monitor of the system. The 4 ms can provide sufficient flash
pulse while allowing for cost effective use of an LED and rolling
shutter accommodation, and allowing for eye safety.
[0079] In some embodiments, a bright light (e.g., approximately 850
nm LED, two LEDs) can be used for the flash pulse. The systems can
therefore have high pulsed current capabilities. The rolling
shutter camera can have a high quality camera lens and NIR
filtering. In some embodiments, the NIR filter can be tighter than
100 nm around the LED center wavelength. In some embodiments, the
system can include optics with a modulation transfer function (MTF)
optimized for the spatial frequencies that enter the system. In
some embodiments, the system can include preprocessing of iris
images for a particular spectrum of spatial frequencies (e.g., low,
medium, high, or the like).
[0080] FIG. 12 is a flowchart illustrating an exemplary process 400
of implementing the biometric analysis systems disclosed herein. To
begin, at step 402, a scene and an object in the scene are
illuminated with dim illumination from one or more illumination
sources. At step 404, the scene is analyzed with an adaptive
trigger module to detect the object (or a region of interest
associated with the object) in the scene during dim illumination.
At step 406, a position in a frame of a rolling shutter camera is
determined with the adaptive trigger module, the position
coinciding with the detected object (or region of interest
associated with the object) in the scene.
[0081] At step 408, a delay is arranged by the adaptive trigger
module between a start of image writing by the rolling shutter
camera and a trigger of the illumination sources such that a stripe
of flash illumination provided by the illumination sources
coincides with the detected object (or region of interest
associated with the object) in the scene. At step 410, movement of
the object is tracked within a field-of-view of the rolling shutter
camera with the adaptive trigger module. At step 412, the delay
between the start of image writing by the rolling shutter camera
and the trigger of the illumination sources is modified with the
adaptive trigger module based on detected movement of the object
(or region of interest associated with the object) out of the
field-of-view.
[0082] FIG. 13 is a flowchart illustrating an exemplary process 500
of implementing the biometric analysis systems disclosed herein. To
begin, at step 502, the adaptive trigger module can set the flash
delay at 0 ms, the flash duration equal to the time of the full
frame, and the flash brightness to low. At step 504, a full frame
image can be acquired with the rolling shutter camera under dim
illumination. At step 506, the adaptive trigger module analyzes the
image captured at step 504 and locates the region of interest in
the image.
[0083] At step 508, the adaptive trigger module sets the flash
delay to a value greater than 0 ms, the flash duration to short
(e.g., a pulse), and the flash brightness to high (e.g., higher
than the dim illumination) to illuminate the region of interest. At
step 510, the adaptive trigger module measures the position of the
object of interest within the region of interest (e.g., eyes of the
user located on the user's face). At step 512, if the object of
interest is within the region of interest, the process 500
continues a loop to constantly (or at predetermined frequencies of
time) ensure that the object is maintained within the region of
interest. At step 514, if the object of interest is near an edge of
the frame or region of interest, the process 500 performs step 516
to correct the flash delay such that the object is centered within
the region of interest. The region of interest can thereby be
maintained in the flash illumination stripe despite relative motion
of the object and/or the camera. In some embodiments, the process
can include a loop that analyzes the captures images to control the
intensity of the flash illumination, thereby avoiding saturation of
the object.
[0084] FIG. 14 is a block diagram of a computing device 600 in
accordance with exemplary embodiments of the present disclosure.
The computing device 600 includes one or more non-transitory
computer-readable media for storing one or more computer-executable
instructions or software for implementing exemplary embodiments.
The non-transitory computer-readable media may include, but are not
limited to, one or more types of hardware memory, non-transitory
tangible media (for example, one or more magnetic storage disks,
one or more optical disks, one or more flash drives), and the like.
For example, memory 606 included in the computing device 600 may
store computer-readable and computer-executable instructions or
software for implementing exemplary embodiments of the present
disclosure (e.g., instructions for operating the illumination
sources, instructions for operating the processing device,
instructions for operating the rolling shutter camera, instructions
for operating the adaptive trigger module, instructions for
operating the communication interface, instructions for operating
the user interface, instructions for operating the central
computing system, instructions for operating the feedback module,
combinations thereof, or the like). The computing device 600 also
includes configurable and/or programmable processor 602 and
associated core 604, and optionally, one or more additional
configurable and/or programmable processor(s) 602' and associated
core(s) 604' (for example, in the case of computer systems having
multiple processors/cores), for executing computer-readable and
computer-executable instructions or software stored in the memory
606 and other programs for controlling system hardware. Processor
602 and processor(s) 602' may each be a single core processor or
multiple core (604 and 604') processor.
[0085] Virtualization may be employed in the computing device 600
so that infrastructure and resources in the computing device 600
may be shared dynamically. A virtual machine 614 may be provided to
handle a process running on multiple processors so that the process
appears to be using only one computing resource rather than
multiple computing resources. Multiple virtual machines may also be
used with one processor. Memory 606 may include a computer system
memory or random access memory, such as DRAM, SRAM, EDO RAM, and
the like. Memory 606 may include other types of memory as well, or
combinations thereof.
[0086] A user may interact with the computing device 600 through a
visual display device 618 (e.g., a personal computer, a mobile
smart device, or the like), such as a computer monitor, which may
display one or more user interfaces 620 (e.g., a graphical user
interface) that may be provided in accordance with exemplary
embodiments. The computing device 600 may include other I/O devices
for receiving input from a user, for example, a camera, a sensor, a
keyboard or any suitable multi-point touch interface 608, a
pointing device 610 (e.g., a mouse). The keyboard 608 and the
pointing device 610 may be coupled to the visual display device
618. The computing device 600 may include other suitable
conventional I/O peripherals.
[0087] The computing device 600 may also include one or more
storage devices 624, such as a hard-drive, CD-ROM, eMMC
(MultiMediaCard), SD (secure digital) card, flash drive,
non-volatile storage media, or other computer readable media, for
storing data and computer-readable instructions and/or software
that implement exemplary embodiments of the biometric analysis
systems described herein. Exemplary storage device 624 may also
store one or more databases 626 for storing any suitable
information required to implement exemplary embodiments. For
example, exemplary storage device 624 can store one or more
databases 626 for storing information, such as data relating to
captured images 126 under dim illumination 104 and flash
illumination 106, regions of interest 120, flash delay 124,
combinations thereof, or the like, and computer-readable
instructions and/or software that implement exemplary embodiments
described herein. The databases 626 may be updated by manually or
automatically at any suitable time to add, delete, and/or update
one or more items in the databases.
[0088] The computing device 600 can include a network interface 612
configured to interface via one or more network devices 622 with
one or more networks, for example, Local Area Network (LAN), Wide
Area Network (WAN) or the Internet through a variety of connections
including, but not limited to, standard telephone lines, LAN or WAN
links (for example, 802.11, T1, T3, 56 kb, X.25), broadband
connections (for example, ISDN, Frame Relay, ATM), wireless
connections, controller area network (CAN), or some combination of
any or all of the above. The network interface 612 may include a
built-in network adapter, network interface card, PCMCIA network
card, PCI/PCIe network adapter, SD adapter, Bluetooth adapter, card
bus network adapter, wireless network adapter, USB network adapter,
modem or any other device suitable for interfacing the computing
device 600 to any type of network capable of communication and
performing the operations described herein. Moreover, the computing
device 600 may be any computer system, such as a workstation,
desktop computer, server, laptop, handheld computer, tablet
computer (e.g., the tablet computer), mobile computing or
communication device (e.g., the smart phone communication device),
an embedded computing platform, or other form of computing or
telecommunications device that is capable of communication and that
has sufficient processor power and memory capacity to perform the
operations described herein.
[0089] The computing device 600 may run any operating system 616,
such as any of the versions of the Microsoft.RTM. Windows.RTM.
operating systems, the different releases of the Unix and Linux
operating systems, any version of the MacOS.RTM. for Macintosh
computers, any embedded operating system, any real-time operating
system, any open source operating system, any proprietary operating
system, or any other operating system capable of running on the
computing device and performing the operations described herein. In
exemplary embodiments, the operating system 616 may be run in
native mode or emulated mode. In an exemplary embodiment, the
operating system 616 may be run on one or more cloud machine
instances.
[0090] FIG. 15 is a block diagram of an exemplary biometric
analysis system environment 700 in accordance with exemplary
embodiments of the present disclosure. The environment 700 can
include servers 702, 704 configured to be in communication with one
or more illumination sources 706, one or more rolling shutter
cameras 708, one or more adaptive trigger modules 710, a feedback
module 712, a user interface 714, and a central computing system
716 via a communication platform 922, which can be any network over
which information can be transmitted between devices
communicatively coupled to the network. For example, the
communication platform 722 can be the Internet, Intranet, virtual
private network (VPN), wide area network (WAN), local area network
(LAN), and the like. In some embodiments, the communication
platform 722 can be part of a cloud environment.
[0091] The environment 700 can include repositories or databases
718, 720, which can be in communication with the servers 702, 904,
as well as the one or more illumination sources 706, one or more
rolling shutter cameras 708, one or more adaptive trigger modules
710, the feedback module 712, the user interface 714, and the
central computing system 716, via the communications platform
722.
[0092] In exemplary embodiments, the servers 702, 704, one or more
illumination sources 706, one or more rolling shutter cameras 708,
one or more adaptive trigger modules 710, the feedback module 712,
the user interface 714, and the central computing system 716 can be
implemented as computing devices (e.g., computing device 600).
Those skilled in the art will recognize that the databases 718, 720
can be incorporated into one or more of the servers 702, 704. In
some embodiments, the database 718 can store data relating to
captured images, regions of interest 120, flash delay 124,
combinations thereof, or the like, can be distributed over multiple
databases 718, 720.
[0093] While exemplary embodiments have been described herein, it
is expressly noted that these embodiments should not be construed
as limiting, but rather that additions and modifications to what is
expressly described herein also are included within the scope of
the invention. Moreover, it is to be understood that the features
of the various embodiments described herein are not mutually
exclusive and can exist in various combinations and permutations,
even if such combinations or permutations are not made express
herein, without departing from the spirit and scope of the
invention.
* * * * *