U.S. patent application number 15/131659 was filed with the patent office on 2016-08-11 for biometric apparatus and method for touch-sensitive devices.
The applicant listed for this patent is Biogy, Inc.. Invention is credited to Waleed Sami Haddad.
Application Number | 20160228039 15/131659 |
Document ID | / |
Family ID | 48082560 |
Filed Date | 2016-08-11 |
United States Patent
Application |
20160228039 |
Kind Code |
A1 |
Haddad; Waleed Sami |
August 11, 2016 |
BIOMETRIC APPARATUS AND METHOD FOR TOUCH-SENSITIVE DEVICES
Abstract
A touch sensor is configured to sense a contact event occurring
between a body part of a user and the touch sensor. The touch
sensor is configured to produce contact data in response to the
sensed contact event. A processor is coupled to the touch sensor
and memory. The processor is configured to store in the memory a
sequence of data frames each comprising contact data associated
with a different portion of the user's body part. The processor is
further configured to generate biometric signature data using the
sequence of data frames.
Inventors: |
Haddad; Waleed Sami; (San
Francisco, CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Biogy, Inc. |
San Francisco |
CA |
US |
|
|
Family ID: |
48082560 |
Appl. No.: |
15/131659 |
Filed: |
April 18, 2016 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
13651408 |
Oct 13, 2012 |
9314193 |
|
|
15131659 |
|
|
|
|
61546838 |
Oct 13, 2011 |
|
|
|
61563138 |
Nov 23, 2011 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H04L 63/0861 20130101;
H04W 12/0605 20190101; A61B 5/117 20130101; G06F 3/011 20130101;
G06F 3/0414 20130101; A61B 5/6898 20130101 |
International
Class: |
A61B 5/117 20060101
A61B005/117; H04L 29/06 20060101 H04L029/06; H04W 12/06 20060101
H04W012/06; G06F 3/041 20060101 G06F003/041; G06F 3/01 20060101
G06F003/01 |
Claims
1. A method, comprising: sensing a contact event occurring between
a body part of a user and a touch sensor; storing, for the contact
event, a sequence of data frames each comprising contact data
associated with a different portion of the user's body part; and
generating biometric signature data using the sequence of data
frames.
2. The method of claim 1, wherein each of the data frames comprises
spatial dimensions corresponding to the geometry of a portion of
the user's body part.
3. The method of claim 1, wherein the sequence of data frames
defines a time-dependent pattern of geometry for different portions
of the user's body part.
4. The method of claim 1, wherein the sequence of data frames
comprises a sequence of geometric images for the contact event.
5. The method of claim 1, wherein each of the data frames
comprises: an involuntary spatial component associated with a
contact pattern that evolves during the contact event; and an
involuntary time component associated with the manner in which the
contact pattern changes over time during the contact event.
6. The method of claim 1, further comprising: measuring pressure
exerted at different locations of the touch sensor by the user's
body part during the contact event; and generating the biometric
signature data using the sequence of data frames and the measured
pressure.
7. The method of claim 1, wherein each frame of data is captured in
response to a triggering event other than a synchronous clock
signal.
8. The method of claim 1, wherein each frame of data is captured in
response to a triggering event, the triggering event comprising
exceeding a predetermined total mass count threshold, the total
mass count corresponding to an area of the touch screen covered by
a portion of the body part at a particular time.
9. The method of claim 1, wherein each frame of data is captured in
response to a triggering event, the triggering event comprising
filling of a designated segment of the touch sensor to a
predetermined level.
10. The method of claim 1, wherein the biometric signature data is
generated during an enrollment process, and the method further
comprises: sensing a subsequent contact event occurring between the
body part of the user and the same or different touch sensor;
storing, for the subsequent contact event, a sequence of subsequent
data frames each comprising contact data associated with a
different portion of the user's body part; generating a contact
pattern using the subsequent sequence of data frames; and
validating or invalidating the contact pattern using the biometric
signature data.
11. The method of claim 1, further comprising: prior to performing
a user-selected function, verifying the biometric signature data of
the user; and performing the user-selected function only upon
successful verification of the biometric signature data.
12. The method of claim 1, wherein the method is performed by a
mobile electronic device.
13. An apparatus, comprising: a touch sensor configured to sense a
contact event occurring between a body part of a user and the touch
sensor, the touch sensor configured to produce contact data in
response to the sensed contact event; and a processor coupled to
the touch sensor and memory, the processor configured to store in
the memory a sequence of data frames each comprising contact data
associated with a different portion of the user's body part, the
processor further configured to generate biometric signature data
using the sequence of data frames.
14. The apparatus of claim 13, wherein each of the data frames
comprises spatial dimensions corresponding to the geometry of a
portion of the user's body part.
15. The apparatus of claim 13, wherein the sequence of data frames
defines: a time-dependent pattern of geometry for different
portions of the user's body part; or a sequence of geometric images
for the contact event.
16. The apparatus of claim 13, wherein each of the data frames
comprises: an involuntary spatial component associated with a
contact pattern that evolves during the contact event; and an
involuntary time component associated with the manner in which the
contact pattern changes over time during the contact event.
17. The apparatus of claim 13, further comprising a pressure sensor
configured to sense pressure resulting from the contact event, the
processor further configured to measure pressure exerted at
different locations of the touch sensor by the user's body part
during the contact event and generate the biometric signature data
using the sequence of data frames and the measured pressure.
18. The apparatus of claim 13, wherein each frame of data is
captured in response to a triggering event other than a synchronous
clock signal.
19. The apparatus of claim 13, wherein each frame of data is
captured in response to a triggering event, the triggering event
comprising exceeding a predetermined total mass count threshold,
the total mass count corresponding to an area of the touch screen
covered by a portion of the body part at a particular time.
20. The apparatus of claim 13, wherein each frame of data is
captured in response to a triggering event, the triggering event
comprising filling of a designated segment of the touch sensor to a
predetermined level.
21-23. (canceled)
Description
RELATED PATENT DOCUMENTS
[0001] This application is a continuation of U.S. patent
application Ser. No. 13/651,408, filed Oct. 13, 2012, which claims
the benefit of Provisional Patent Application Ser. Nos. 61/546,838,
filed on Oct. 13, 2011, and 61/563,138, filed on Nov. 23, 2011, to
which priority is claimed and which are hereby incorporated herein
by reference.
SUMMARY
[0002] Embodiments of the disclosure are directed to a biometric
system and method for use in devices that include a touch sensor.
Method embodiments of the disclosure involve sensing a contact
event occurring between a body part of a user and a touch sensor
and storing, for the contact event, a sequence of data frames each
comprising contact data associated with a different portion of the
user's body part. Method embodiments further involve generating
biometric signature data using the sequence of data frames.
[0003] Apparatus embodiments of the disclosure include a touch
sensor configured to sense a contact event occurring between a body
part of the user and the touch sensor. The touch sensor is
configured to produce contact data in response to the sensed
contact event. A processor is coupled to the touch sensor and
memory. The processor is configured to store in the memory a
sequence of data frames each comprising contact data associated
with a different portion of the user's body part. The processor is
further configured to generate biometric signature data using the
sequence of data frames.
[0004] These and other features can be understood in view of the
following detailed discussion and the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
[0005] FIGS. 1-5 are flow diagrams showing biometric processes in
accordance with various embodiments of the disclosure;
[0006] FIG. 6 is a flow diagram showing biometric processes
including generation of visual and/or audio instructions in
accordance with various embodiments of the disclosure;
[0007] FIG. 7 is a flow diagram showing biometric processes
involving an enrollment process for generating a biometric
signature and post-generation use of the biometric signature in
accordance with various embodiments of the disclosure;
[0008] FIG. 8 is a flow diagram showing biometric processes
involving unlocking of an electronic device using a biometric
signature in accordance with various embodiments of the
disclosure;
[0009] FIGS. 9-11 and 13 are flow diagrams showing biometric
processes involving a triggering algorithm in accordance with
various embodiments of the disclosure;
[0010] FIGS. 12 and 14 are diagrams depicting aspects of biometric
processes involving the triggering algorithms of FIGS. 11 and 13,
respectively, in accordance with various embodiments of the
disclosure; and
[0011] FIG. 15 is a block diagram of a system which includes an
electronic device configured to implement a biometric system and
method according to various embodiments of the disclosure.
DESCRIPTION
[0012] In the following description of the illustrated embodiments,
references are made to the accompanying drawings forming a part
hereof, and in which are shown by way of illustration, various
embodiments by which the invention may be practiced. It is to be
understood that other embodiments may be utilized, and structural
and functional changes may be made without departing from the scope
of the present invention.
[0013] Embodiments of the disclosure relate to a biometric system
and method for use with a wide variety of systems and devices. A
biometric system and method according to embodiments of the
disclosure find particular utility when incorporated into mobile
electronic devices, such as portable communication devices.
[0014] In the context of portable communication devices, there is a
rapidly increasing need for a biometric system that can be
implementable on common, currently available mobile devices, mainly
mobile phones and tablet computers. This need appears to be
becoming critical as the use of mobile devices for conducting
transactions, and in particular financial transactions, increases,
while the sophistication, and therefore the risk, of cyber crime
grows.
[0015] Despite the apparent trajectory for the use of mobile
devices, and the obvious advantages of biometrics as a component of
an effective security protocol, there is no readily available
biometric device that is easily implementable on mobile phones.
There are small, inexpensive fingerprint readers (swipe sensors,
and a few area sensors) that are appearing on some phone models, as
well as a few voice and face recognition applications that can be
used with the microphone or camera available on most phones,
however these work poorly, and would require additional, or much
better biometric hardware to be added to the phone in order for
them to be reliable and accurate enough to provide serious
security. User acceptance is also a very important factor in
biometrics, and many users are wary of the familiar biometric
systems, and in particular of fingerprint readers.
[0016] A new biometric system that is easy to use, simple to
implement using existing hardware on mobile phones and tablets, and
is different enough from any of the common biometrics that are used
now could fill the need for an effective way of verifying a user's
identity without a long technology development cycle, or the need
for hardware manufacturers to build a new device into their
products. This would allow the added security of biometrics to be
put in place on the very short timescale needed to support the
rising security needs of mobile transactions.
[0017] Embodiments of the inventions described herein use modern
touch screen technology to acquire the patterns of one, two, three
or four of the user's fingers pressed onto the screen together, or
in a particular sequence. To increase the sophistication, and
amount of usable biometric information, a time-dependent pattern is
used, not just a simple static finger geometry pattern. In other
words, a rapid series of "frames" of the user's finger geometry is
recorded as he/she presses down onto the screen. Because of the
natural 3-D geometry of a user's fingers and hand, this
time-dependent pattern has additional and useful complexity. This
time-dependence actually allows each user to develop a special
technique of hand placement, perhaps with additional conscious
movements that only he/she knows, in a sense adding a password-like
component to the pattern.
[0018] Since the user's hand geometry is unique, and not controlled
by the user, there will always be involuntary components to the
biometric signature that are unknown, even to the user, both
spatially (the hand geometry pattern in 2-D on the screen), as well
as the involuntary time dependence (the way the pattern changes
over time as the user presses down on the screen). This is critical
for biometric security because it cannot be given away, even by the
user himself/herself.
[0019] The additional possibility that the user can also make extra
voluntary movements during hand placement creates even more
uniqueness, and therefore valuable complexity to the biometric
"signature" that will be used by the system. Furthermore, the
approach is not limited to hand or finger geometry. It can be used
with other body parts, most notably the pinna of the ear. This can
be seen as potentially desirable since most touch screen phones are
constructed in such a way that when the user holds the phone to
talk, he/she will inevitably press his/her ear against the touch
screen somewhat consistently every time the phone is used. It may
be possible to successfully enroll the ear print pattern in the
same way as the time-dependent hand print pattern would be, and
then to use this ear print pattern as another biometric signature.
In principle, other body parts could also be used, such as the
lips, knuckles, a part of the arms or legs, foot, etc. These are
not likely to be as practical as the hand, or even the pinna of the
ear, but they are mentioned here for completeness.
[0020] One significant advantage of the added element of
time-dependence of the biometric signature is that there will not
be any possibility for hackers to "lift" a so-called "latent
print." Latent prints are a problem with some fingerprint systems
because the user may leave a clear image of his/her fingerprint on
the sensor (or possibly elsewhere) that can be copied using one of
several methods, and used to create a false fingerprint, or, in
some cases, the latent print left on a sensor can be induced to
trigger the sensor again through the application of heat, cold,
illumination, or something else, depending on the sensor
technology. The issue of latent prints is essentially rendered
inconsequential by the element of time-dependence in the approaches
described herein.
[0021] This approach is chosen because the resolution of current
touch screens is suitable for such use, but they are not capable of
resolving finer patterns, such as those of a fingerprint. The use
of time dependence in the biometric embodiments disclosed herein
represents a unique capability not found in conventional
biometrics. The temporal component of the data differentiates it
from other biometrics, and opens up a number of possibilities for
the user to participate in creating the "password" or "key", as
well as for increasing the complexity of the biometric
signature.
[0022] The spatial dimension of the data will be quite
low-resolution with current state-of-the-art touchscreens, and
therefore not by itself contain sufficient information, however,
with the time dimension added, the dataset will effectively be
three-dimensional, and contain a great deal of information that can
be used to differentiate a large number of users, as well as
provide enough information density to overcome the difficulties of
natural variations in biometric signatures from which all biometric
systems suffer. If the touchscreen happens to have the ability to
also measure pressure as a function of position on the screen, such
as a pressure-sensitive touch screen does, then yet another
dimension can be added to the biometric signature: that of
pressure. The data set would now effectively be four-dimensional,
having two spatial dimensions and pressure, all as a function of
time.
[0023] Turning now to FIG. 1, there is illustrated a flow diagram
showing biometric processes in accordance with various embodiments.
The biometric method depicted in FIG. 1 involves sensing 110 a
contact event involving a body part of a user, and producing 112
contact data. The contact data preferably includes spatial data
associated with the 2-D geometry of the user's body part and
temporal data associated with the development of a contact pattern
for the contact event. The method of FIG. 1 further involves
generating 114 biometric signature data using the contact data. It
is noted that content of the contact data can be different for
different embodiments of the disclosure.
[0024] A typical contact event involves an intentional touching of
a touch sensitive device by a user. For example, the user may place
one or more fingers (or palm, for example) on a touch sensor of the
touch sensitive device, which can define a contact event. By way of
further example, the user may use one or more fingers to swipe
across a region of the touch sensor, which can define a contact
event. It is understood that a wide variety of static (stationary)
and dynamic (moving) contact events are contemplated. It is noted
that, in the case of a static contact event, a resulting contact
event still involves development of a contact pattern over time,
since the area of contact between the user's body part and the
touch sensor changes between initial contact and a stationary
state.
[0025] FIG. 2 illustrates a flow diagram showing various biometric
processes in accordance with other embodiments. In FIG. 2, the
biometric method involves sensing 120 a contact event involving a
body part of a user, and producing 122 contact data. The contact
data preferably includes data indicative of a time-dependent
pattern of the geometry of different portions of the user's body
part as the contact event evolves over time. The method of FIG. 2
further involves generating 124 biometric signature data using the
contact data.
[0026] The flow diagram illustrated in FIG. 3 shows various
biometric processes in accordance with some embodiments. The
biometric method shown in FIG. 3 involves sensing 130 a contact
event involving a body part of a user, and producing 132 contact
data. The contact data according to this embodiment preferably
includes data indicative of a chronology of frames of data of the
contact event (referred to herein as "contact data frames"). The
method of FIG. 3 further involves generating 134 biometric
signature data using the contact data.
[0027] FIG. 4 illustrates a flow diagram showing various biometric
processes in accordance with further embodiments. In FIG. 4, the
biometric method involves sensing 140 a contact event involving a
body part of a user, and producing 142 a sequence of data frames
each comprising contact data associated with a different portion of
the user's body part as the contact pattern develops. The method of
FIG. 4 also involves generating 144 biometric signature data using
the contact data.
[0028] Embodiments of the disclosure can acquire other and/or
additional data for purposes of implementing various biometric
processes. According to the embodiment shown in FIG. 5, the
biometric method involves sensing 150 a contact event involving a
body part of a user, and sensing 152 pressure resulting from the
contact event. Contact data is produced 154 for this contact event
from which a contact pattern can be developed. The contact data in
FIG. 5 preferably includes spatial data associated with the 2-D
geometry of the user's body part, pressure data associated with the
touch, and temporal data associated with the development of a
contact and pressure profile for the contact event. The method of
FIG. 5 also involves generating 154 biometric signature data using
the contact data.
[0029] FIG. 6 illustrates a flow diagram showing various biometric
processes in accordance with some embodiments. In FIG. 6, the
biometric method involves generating 160 visual and/or audio
instructions to aid a user in creating or validating a biometric
signature. The method also involves sensing 162 a contact event
involving a body part of a user in accordance with the generated
instructions, and producing 164 contact data indicative of spatial
dimensions of the user's body part as a function of time. The
method of FIG. 6 further involves generating 166 biometric
signature data using the contact data.
[0030] The flow diagram of FIG. 7 illustrates an enrollment process
for generating a biometric signature and post-generation use of the
biometric signature in accordance with various embodiments of the
disclosure. During the enrollment process, a user establishes his
or her biometric signature which, once created, can be used to
provide secured access to electronic devices, applications,
websites, and other secured systems and software processes (e.g.,
bank transactions via a mobile phone or tablet).
[0031] The enrollment process shown in FIG. 7 (blocks 170-175)
involves the user placing 170 a body part on a touch sensor. The
user may select which body part (e.g., hand or ear) and number of
body parts to be used to create the user's biometric signature. For
example, the user may choose to use 3 figures that are placed near
the center of the touch sensor. In another example, the touch
sensor may be pressed against the pinna of the user's ear, which
may be convenient for secured use of mobile phones that incorporate
a touch sensor.
[0032] With the body part being placed on the touch sensor, the
enrollment process involves sensing 171 a contact event involving
the selected body part(s), and producing 172 contact data. The
contact data according to this embodiment preferably includes data
indicative of a time-dependent pattern of the geometry of different
portions of the user's body part as the contact event evolves over
time. The method further involves generating 173 a biometric
signature for the user using the contact data. The processes of
blocks 170-173 may be repeated 174 to enhance the reliability
(e.g., stability, repeatability) of the user's biometric signature.
The enrollment process concludes with storing 175 the user's
biometric signature for subsequent use. The biometric signature can
be stored locally on a mobile electronic device owned by the user,
on a remote server or both locally and remotely. The user's
biometric signature is now available for use with various secured
applications, websites, services, systems, and devices that require
user authentication.
[0033] In the post-enrollment use example illustrated in FIG. 7
(blocks 176-182), it is assumed that a biometric algorithm of the
present disclosure is operating on an electronic device and that
the device is operating an application that requires user
authentication prior to allowing use or access to the device and/or
application. At a time subsequent to the enrollment process, the
method of FIG. 7 further involves attempting 176 to execute a
secured application on the electronic device, which may be a mobile
phone, tablet or PC, for example. Attempting to execute the
application involves creating a contact pattern for the user in the
same manner as when generating the user's biometric signature. It
is understood that the touch sensor and/or electronic device used
to generate the biometric signature can be the same as, or
different from, those used to produce the contact pattern.
[0034] With continued reference to FIG. 7, a touch sensor of the
electronic device senses 177 contact with a body part of the user
and produces 178 contact data in a manner previously described. The
body part(s) brought into contact with the touch sensor of the
electronic device is the same as that/those used to create the
user's biometric signature. The just-produced contact pattern is
compared 180 to the user's biometric signature to authenticate the
user. This comparison can be performed by the electronic device, by
a remote system communicatively coupled to the electronic device,
or cooperatively by both entities. The user is granted access 182
to the application only if the contact pattern matches the
biometric signature.
[0035] FIG. 8 illustrates a flow diagram showing various biometric
processes in accordance with some embodiments. In FIG. 8, a
procedure for unlocking 190 an electronic device is initiated. It
is assumed in the illustrative embodiment of FIG. 8 that a
biometric algorithm of the present disclosure is operating on the
electronic device and that the device is operating an application
that requires user authentication prior to allowing use or access
to the device and/or application. The unlocking procedure involves
interrupting 192 a normal password prompt typically generated by
the application. Instead, a biometric signature verification
procedure of the present disclosure is initiated 194. Contact data
resulting from sensing contact between a user's body part and a
touch sensor of the electronic device is acquired, from which a
time-dependent contact pattern is generated. This time-dependent
contact pattern is used by the biometric algorithm to verify 196
authenticity of the user.
[0036] The biometric algorithm passes 198 back control to the
application and the electronic device is unlocked only if the
verification procedure is successful. If the verification procedure
is unsuccessful (i.e., the contact pattern constructed from the
contact data does not match the user's biometric signature), a
signal indicative of such failure is generated 200, and the
electronic device is maintained in the locked state. In response to
the generated signal, a message indicating the unsuccessful
verification is communicated to the user, typically via a visual
and/or audio message.
[0037] In accordance with the biometric processes shown in FIG. 9,
various embodiments of the disclosure provide for detection 201 of
a triggering event which, when detected, causes initiation of
biometric signature data acquisition. Detection 201 of a triggering
event causes initiation 202 of a procedure for detecting a contact
pattern of a user's body part relative to a touch sensor that
results from a contact event with the touch sensor. The method
further involves recording 204 contact pattern data comprising
spatial dimensions of the user's body part in contact with the
touch sensor as a function of time. The method also involves
terminating 206 the detection and recording processes after either
successful completion or failure of the contact pattern detection
procedure. It is understood that a variety of triggering
methodologies are contemplated, and that a triggering protocol can
be implemented for one or both of the biometric signature and
contact pattern generation processes.
[0038] The addition of the time dimension adds a complexity that is
both desirable, and potentially excessive, however this complexity
can be quantized and controlled through creative design of the data
acquisition algorithms. The excessive complexity can come from the
fact that a user may change the "speed" with which he/she executes
the hand placement on the touchscreen, effectively shortening, or
lengthening the extent of the dataset in the time dimension
depending on if the placement is faster or slower. This would be
the case if acquisition of each "frame" of the pattern is acquired
according to an independent clock that runs on the device,
collecting the data in fixed time steps in the same way a movie
camera, for example, acquires a sequence of images of a scene. This
can make matching with the enrolled pattern difficult. One way to
handle this is to develop algorithms that can compress or stretch
the dataset in the time dimension as part of the pattern-matching
component of the biometric algorithm suite. Such a method may be
based on existing "morphing" techniques, wavelet transforms, or
other known methods, or it may be developed specifically for use in
this application. This is one viable approach, however, it is not
likely to be the fastest or most efficient, and may have other
problems.
[0039] An alternate, and possibly better approach would be to
quantize the time dimension of the data during acquisition based on
the pattern itself, and not on an independent clock. This would
require specialized triggering algorithms to "step" the acquisition
of the pattern on the screen as a function of time while the user
is making the placement. Since the only important thing in the time
dimension is what aspects of the pattern appear first, second,
third, and so on, by quantizing time through special triggering
that is based on the fundamental properties of the pattern itself
as it develops during the placement, the time axis of the data set
will be controlled in real time, during the placement
unintentionally by the user him/herself in an automated fashion.
The time aspect of the user's hand placement need not be linear in
this case, as the triggering algorithm will "examine" the pattern
as it develops in time, and trigger the capture of the "movie
frames" of the pattern automatically based on some criteria of the
image (pattern) itself. There are several possible designs for this
specialized triggering algorithm, examples of which are described
below.
[0040] FIG. 10 illustrates a flow diagram showing various biometric
processes including a triggering methodology in accordance with
various embodiments. FIG. 10 shows a number of different processes
involving generating a contact pattern in response to a triggering
event and validating this contact pattern against a pre-established
biometric signature for the user. In general terms, the triggering
processes described herein rely on storing of a sequence of data
"points", which are taken during the enrollment process or are
predetermined somewhat arbitrarily. During use of a secured
electronic device, for example, when the user places his/her hand
on the touch sensor (e.g., screen), the thresholds for triggering
are based on these stored data points (for example, a series of
different total mass count levels as is described below). As the
placement occurs, the decision to trigger capturing of a frame of
contact data is made based on the first data point (ith data
point), then, once that is done, the next triggering event is based
on the second data point (ith+1 data point), and so forth. As
discussed previously, the triggering and contact data generation
methodologies shown in FIG. 10 can be implemented for one or both
of the biometric signature and contact pattern generation
processes.
[0041] The processes shown in FIG. 10 include sensing 210 a contact
pattern evolving in real-time for a contact event involving a body
part of a user. The method also involves acquiring 212 data from
the contact event that is used by a triggering algorithm, which is
typically running on an electronic device operable by the user. If
the acquired data meets or exceeds a triggering threshold 214, a
triggering event is declared and a frame of contact data is
captured 216 at a current clock time Tx. If the acquired data fails
to meet or exceed the triggering threshold 214, the data
acquisition process of block 212 continues.
[0042] Each triggering event 214 results in the capturing 216 of an
additional frame of contact data for the contact event. Over time,
the sequence of captured data frames defines a contact pattern that
evolves as the contact event evolves. At some stage of the
procedure, a validation operation occurs to verify whether or not
the developing or developed contact pattern corresponds to the
pre-established biometric signature of the user. In some
embodiments, each captured contact data frame is added 218 to a
developing contact pattern which, when sufficiently formed, is
subsequently subjected to validation. In other embodiments, each
captured frame of contact data to be added to a developing contact
pattern is subjected to validation 220 against the pre-established
biometric signature of the user. By way of example, if a given
frame of captured contact data is determined to be out of sequence
relative to its expected position within the contact data frame
sequence of the biometric signature, this contact data frame would
be considered invalid.
[0043] As the contact data capturing routine continues, a test 224
is made to determine if enough contact data has been collected for
the developing contact pattern. If not, the clock continues to run
228 and the data acquisition process in block 212 continues. If a
sufficient amount of contact data has been collected for the
developing contact pattern, the contact pattern is compared 226 to
the user's pre-established biometric signature. If determined to be
invalid, the procedure of FIG. 10 is terminated 232 and an
invalidity signal is generated. If determined valid, the procedure
of FIG. 10 is terminated 234 and a signal confirming validity is
generated. It is understood that the validation processes of blocks
226, 230, and 232 would not be implicated when creating the user's
biometric signature in accordance with the methodology shown in
FIG. 10.
[0044] Verification of a biometric signature can involve a number
of different validation techniques. One approach involves comparing
a temporal order of data frames of a developing or developed
contact pattern to that of the biometric signature. Another
approach involves comparing characteristics of the time-dependent
spatial data of a contact pattern with those of the biometric
signature. It is to be understood that there are many ways to use
the contact information collected in space and time, including
arranging the 3-D data (2-D in space, and 1-D in time) into a new
format of 2-D data that can be tested for verification using
existing pattern recognition methods. In fact, the biometric
signature can be created from this data in a number of different
ways.
[0045] According to some embodiments, triggering of screen image
capture can be based on "total mass" count (TMC) of the image as a
function of time. Triggering can be quantized based on the total
amount of "mass" or ratio of bright to dark pixels of the screen
that are filled. As the user places his/her hand on the
touchscreen, the area filled by the parts of the screen that are
covered by the hand is added up to provide a "total mass" count
(TMC) as a function of time. This TMC can be based on either a
binarized version of the image, or the grayscale version. It may be
the case that using the binarized version of the image to calculate
the TMC will be more reliable for triggering purposes, but this
will depend on the properties of the screen, and the user's
hand.
[0046] Triggering on the capture of image data of the user's hand
during placement will then be set to occur upon reaching specific
values of the TMC in sequence as the user places his/her hand on
the screen during verification. Thus, triggering of sequential
"frames" of the touchscreen image of the user's hand will not
depend on the time, but upon the image data itself being captured.
The TMC values used for the triggering can be simply arbitrary
values from low to high, or a set of other, perhaps non-linearly
increasing values of the TMC. The set of TMC triggering values may
be determined at the time of enrollment by the user, and can be
based on the TMC values reached over a linear time sequence during
the enrollment process, after which the originally-used time
sequence can be abandoned, and triggering will be based on the
predetermined TMC values. Otherwise, any other favorable sequence
of TMC values can be predetermined, and used for the triggering
process, if such a set of values is known from testing to deliver
reliable triggering results over a variety of users.
[0047] FIG. 11 illustrates a flow diagram showing various biometric
processes including a triggering methodology in accordance with
some embodiments. In FIG. 11, the biometric method involves sensing
240 a contact pattern evolving in real-time for a contact event
involving a user's body part. While the contact pattern continues
to evolve 242 (e.g., for i=1 to N), the total mass count (ith TMC)
of the contact pattern image is determined 244. If the ith TMC
exceeds an ith predetermined threshold 246, a triggering event is
declared 250, and an ith frame of contact data is captured 252 at
the current clock time Tx. If not, the clock continues to run and
the processes beginning at block 242 continue. The processes above
continue 254 until a sufficient amount of contact data (e.g., i=N)
has been captured, at which point a validation process may be
initiated. FIG. 12 illustrates an evolving contact pattern relative
to predetermined triggering thresholds. When the ith TMC of the
contact pattern exceeds the ith triggering threshold, an ith frame
of contact data is captured at the current clock time Tx, thereby
creating a sequence (i=1 to N) of contact data frames.
[0048] According to other embodiments, triggering of screen image
capture can be based on sequential filling of spatial regions of
the screen. Instead of using TMC to determine the points in time at
which each new image frame is captured during the hand placement on
the screen, the filling of various regions of the screen can be
used as triggering events. This can be done by dividing up the
screen area into a number or regions prior to verification. This
can be done arbitrarily, for example by dividing the area up on a
2-D grid, with rectangular regions that are equal to, or larger
than, the fundamental spatial resolution of the screen, or using
any other arbitrary, pre-determined segmentation desired, such as
concentric arc-shaped segments, etc. The arbitrary segmentation can
be done, and stored in the device memory prior to the use of the
application by any user, or prior to its use by each user. It can
also be done after enrollment by the user, thereby making the
segmentation map unique for each user.
[0049] By choosing the shape and size of segments of the screen to
examine during later verification attempts based on how the user
placed his/her fingers during enrollment, the triggering scheme can
be optimized for each user, and for the way each user places
his/her fingers on the screen. This would be done via an algorithm
that analyzes the finger contact pattern during enrollment, and
storing information about the time sequence in which various
regions of the screen are "filled" as the user enrolls. The exact
time at which a region is filled does not matter, only the relative
time at which regions get filled.
[0050] During verification, first, a start command would initiate
the process, and the device will wait to capture a full screen
image until the first triggering region is filled. Once this
happens, the device again waits until the next screen region in the
sequence is filled, and triggers another full-screen image capture.
This process continues until verification is terminated. The
termination point can also be triggered either by the filling of
all the designated regions, or when the final region in the
sequence is filled. It can also be terminated early if the
designated regions are filled out of sequence. This can be used as
part of the verification process.
[0051] A user's verification attempt can be rejected early if the
placement fills the regions in a sequence that is too different
from the sequence (or sequences, since enrollment will usually
require more than one placement for reliability) stored during
enrollment. If the user has shifted his/her position up, down, left
or right along the screen during verification compared with the
original location during enrollment, a simple translation shift can
be applied when analyzing the filling of the screen regions in
order to compensate for this. A similar sort translation shift must
be used when comparing the actual enrollment pattern with any
verification pattern as well.
[0052] FIG. 13 illustrates a flow diagram showing various biometric
processes including a triggering methodology in accordance with
some embodiments. In FIG. 13, the biometric method involves sensing
260 a contact pattern evolving in real-time for a contact event
involving a user's body part. While the contact pattern continues
to evolve 262 (e.g., for i=1+N), the filling of designated touch
sensor segments due to being covered by the user's body part is
examined 264. A test is made to determine 266 if a current (ith)
segment is properly filled relative to an predetermined ith
threshold. If not, the clock continues to run 268 and the processes
beginning at block 262 continue. If the current segment is properly
filled, a triggering event is declared 270 and an ith frame of
contact data is captured 272 at the current clock time Tx, thereby
creating a sequence of contact data frames. The processes above
continue 274 until a sufficient amount of contact data has been
captured, at which point a validation process may be initiated.
FIG. 14 illustrates an evolving contact pattern relative to a
predetermined triggering thresholds. When filling of an ith segment
during contact pattern evolution exceeds an ith triggering
threshold, an ith frame of contact data is captured at the current
clock time Tx, thereby creating a sequence of contact data frames
(i=1 to N).
[0053] FIG. 15 is a block diagram of a system 300 which includes an
electronic device configured to implement a biometric system and
method according to various embodiments of the disclosure. The
system 300 shown in FIG. 15 includes a touch sensor 302 coupled to
a processor 304. The touch sensor 302 may include one or more
user-actuatable buttons 303. The touch sensor may be fabricated
according to various technologies, including capacitive, resistive,
force, and acoustic technologies, for example. The touch sensor 302
may optionally incorporate an integral or separate pressure sensor
capable of sensing pressure at various locations of the sensor
surface. The processor 304 includes a clock 305 which may be
separate from the main clocks of the processor 304. The clock 305
may be dedicated to perform clocking functions for the biometric
signature algorithms 307 stored in a memory 306 coupled to the
processor 304. A speaker and/or microphone unit 308 may be
included.
[0054] The system 300 may further include one or more wired and/or
wireless communication units 310, such as one or more radios (e.g.,
cellular, Wi-Fi), transceivers (e.g., Bluetooth), and hardwire
interfaces (e.g., Ethernet). The communication unit(s) 310 are
coupled to the processor 304 and provide communicate coupling to
external systems and networks, such as the Internet 312. The
processor 304 may communicate with a remote server 314, for
example, via the Internet 312 or other communication link. As
discussed previously, biometric data can be transferred between the
processor 304/memory 306 and the remote server 314. The processor
304 and the remote server 314 may operate cooperatively during one
or more biometric processes described hereinabove. Components of
the system 300 shown in FIG. 15 can be incorporated in a variety of
electronic devices, such as a mobile phone, tablet, PC, and the
like.
[0055] The foregoing description of the example embodiments has
been presented for the purposes of illustration and description. It
is not intended to be exhaustive or to limit the inventive concepts
to the precise form disclosed. Many modifications and variations
are possible in light of the above teaching. Any or all features of
the disclosed embodiments can be applied individually or in any
combination are not meant to be limiting, but purely illustrative.
It is intended that the scope be limited not with this detailed
description, but rather determined by the claims appended
hereto.
* * * * *