 |
 |
 |
|
FingerCell Source Code EDK
Embedded and mobile fingerprint identification
technology
 Click
Here
to
download
the
full
VeriFinger
SDK
brochure
as a .pdf document
The FingerCell technology, developed on the VeriFinger basis, is
designed for embedded biometric systems developers. FingerCell
algorithm is compact, sensor-independent and cross-platform. It offers
decent reliability and identification speed for various mobile or
embedded devices
FingerCell is available for integrators as Embedded Development Kits
(EDK) with FingerCell library or source code for developing a fast and
reliable system on embedded or mobile platform.
Advantages of FingerCell
- Reliability proven at FVC2004
- Multiple fingerprint sensors support
- Cross platform algorithm with compact portable source code.
- Low speed processors supported.
- Reasonable prices, flexible licensing and free customer support.
FingerCell Technology and EDK
- FingerCell Embedded Development Kits. FingerCell
is available for integrators as 2 types of embedded development kits. FingerCell
Library
EDK is intended for biometric system projects using
hardware based on ARM processors. FingerCell source code EDK
is intended for large biometric system projects using third party or
custom hardware… Read more
- System requirements. Recommended processor is
at least 200 MHz ARM family CPU, and minimum required
processor is 75 MHz ARM7 CPU. At least 400 kB
of memory required for algorithm code and data arrays. Windows
CE and ARM-Linux are supported… Read more
- Reliability testing results and technical specifications.
FingerCell algorithm enrolls a fingerprint in less than 1
second, matches up to 700 fingerprints per second
and requires 300-600 bytes of memory to store a
fingerprint template… Read more
- Download. FingerCell brochure, FingerCell algorithm demo applications for
Windows CE and Win32 platforms, and FingerCell
EDK
30-day
Trial are available for downloading.
- References. Neurotechnology customers have
developed a number of embedded and mobile solutions based on FingerCell
software. Please contact
us today to learn more.
Back to top of this page
Contents of FingerCell Library and Source EDK
FingerCell Embedded Development Kit (EDK) is based on the FingerCell
embedded fingerprint
recognition algorithm that is especially designed to be used in
embedded low-power and comparably low-CPU-power applications.
FingerCell EDK includes libraries for ARM Linux and ARM Windows CE
embedded platforms, as well as support modules for embedded sensors and
source codes for sample applications.
Customers who want to use another platform can obtain the FingerCell
ANSI C source code package and port
the software to the required platform.
The following types of FingerCell 2.1 EDK are available:
- FingerCell 2.1 Library EDK – is intended
for biometric system projects using hardware based on ARM processors.
Includes FingerCell library, programming samples and documentation for
Windows CE and Linux.
- FingerCell 2.1 source code EDK – is
intended for large biometric system projects using third party or
custom hardware. Includes FingerCell source code, samples and
documentation for MS Windows CE and Linux.
The table below compares different types of FingerCell EDK:
| |
Library EDK |
Source code EDK |
| Supported
platforms |
| ARM, Windows CE |
+ |
+ |
| ARM, Linux |
+ |
+ |
| FingerCell
algorithm
components |
| • FingerCell 2.1 algorithm |
+ |
+ |
| • FingerCell 2.1 algorithm source code |
|
+ |
| Scanner
support
modules
(for
Linux) |
| • Tacoma CMOS scanner support module |
+ |
+ |
| • Startek FM200 scanner support module |
+ |
+ |
| • Biometri-CS CS-Pass sensor support
module |
+ |
+ |
| • Zvetco Verifi P4000 scanner support
module |
+ |
+ |
| • AuthenTec AF-S2 sensor support module |
+ |
+ |
| • AuthenTec AES4000 sensor support
module |
+ |
+ |
| • Fujitsu MBF200 scanner support module |
+ |
+ |
| FingerCell
programming
samples |
| • FingerCell EDK sample application |
+ |
+ |
| Documentation |
| • FingerCell EDK documentation |
+ |
+ |
| • FingerCell source code documentation |
|
+ |
FingerCell 2.1 Library EDK
FingerCell 2.1 Library EDK includes the FingerCell 2.1 library for
developing custom products. The developed product can run on ARM-based
platform under Linux or Microsoft Windows CE.
FingerCell 2.1 Library EDK contains the following
components:
- MS Windows CE components:
- FingerCell 2.1 library (for Microsoft Visual Studio 2005 with
SP1)
- Source code of FingerCell library usage sample application in
Visual C++ 2005 SP1
- ARM Linux components:
- FingerCell 2.1 library (for Arm-Linux GCC C compiler)
- Source code of sample embedded application in ANSI C (project
for Arm-Linux GCC C compiler)
- User-space drivers for image input from Tacoma CMOS, Startek
FM200, Biometri-CS CS-Pass, Zvetco Verifi P4000, AuthenTec AF-S2,
AuthenTec AES4000 and Fujitsu MBF200 fingerprint sensors via USB port
- FingerCell 2.1 EDK documentation.
FingerCell 2.1 source code EDK
FingerCell 2.1 source code EDK is intended for developers who are going
to integrate fingerprint identification technology into a custom
embedded device.
FingerCell 2.1 source code EDK contains the
following components:
- 10,000 FingerCell 2.1 installation licenses
- FingerCell 2.1 source code:
- Project for GCC compiler (ARM-Linux platform)
- Project for MS Visual Studio 2005 (Pocket PC 2003 platform)
- FingerCell 2.1 Algorithm and Source Code Description
- Sample applications:
- Project for GCC compiler (ARM-Linux platform)
- Project for MS Visual Studio 2005 (Pocket PC 2003 platform)
- Linux user-space drivers' source codes for Tacoma CMOS, Startek
FM200, Biometri-CS CS-Pass, Zvetco Verifi P4000, AuthenTec AF-S2,
AuthenTec AES4000 and Fujitsu MBF200 fingerprint sensors connected via
USB port
- FingerCell 2.1 EDK developers' guide
Back to top of this page
FingerCell EDK System Requirements
- ARM-based processor:
- Minimum requirement: ARM7 processor running
at 75 MHz for fingerprint verification in about 2 seconds.
- Recommended: ARM-based processor with 200
MHz CPU clock rate for fingerprint enrollment in less than 1 second
(ARM processor core families: ARM9, ARM10, ARM11, StrongArm, XScale).
- At least 400 KB of memory for FingerCell code
and data arrays (the recommended amount could be different, as it
depends on fingerprint image size)
- Fingerprint sensor that has a support module included in
FingerCell EDK or a driver is available from scanner manufacturer or
other sources
- ARM Linux (glibc 2.3.4 or later) or Microsoft Windows
Mobile
2003 (or later) operating system
Please note that FingerCell 2.1 source code EDK can be easily ported
to most other platforms and processors using ANSI C compiler.
Back to top of this page
Reliability and Performance Test Results
All tests were
performed on Intel Core2 processor with 4 cores running at 2.66 GHz.
As FingerCell is intended for embedded devices, it uses a faster and
less powerful fingerprint noise filtration algorithm with a slightly
higher False Rejection Rate than a PC running the VeriFinger algorithm.
However, the FingerCell algorithm still produces a decent level of
recognition reliability, which is acceptable for embedded devices.
Receiver operation characteristic (ROC) curves obtained in testing with
two scanner databases compare FingerCell 2.1 (green) and VeriFinger 6.0
(red) reliability under the same conditions.
All fingerprint templates should be loaded into RAM before
identification, thus the maximum fingerprint templates database size is
limited by the amount of available RAM.
| FingerCell
2.1
algorithm
technical
specifications |
| Enrollment time |
< 1 second |
| Enrollment time in features generalization
mode |
< 3 seconds |
| Verification time |
0.5 seconds |
| Matching speed |
up to 700 fingerprints/second |
| Template size |
300 - 600 bytes |
| Memory required for code and data arrays |
400 kilobytes |
Back to top of this page
|
|
|
 |
|
 |
|
|
|
 |
|
 |
|
FingerCell Algorithm Features and Capabilities
All performance tests were made on
Intel Core2 processor with 4 cores running at 2.66 GHz.
FingerCell offers decent reliability and identification speed for
various mobile or embedded devices. The FingerCell algorithm is similar
to the VeriFinger algorithm and includes these proprietary solutions:
- Fast image processing. Fingerprint image
processing time is less than 1 second on 200 MHz ARM
processor, which is acceptable for embedded systems.
- Low speed processors are supported. The template
extraction and matching is adapted for low speed embedded processors.
For example, fingerprint verification is performed in about 2
seconds on a 75 MHz ARM7 processor when FingerCell algorithm
is used.
- Features generalization mode. This fingerprint
enrollent mode generates a collection of the generalized fingerprint
features from a collection of fingerprints of the same finger. Each
fingerprint image is processed and features are extracted. Then the
collection of features is analyzed and combined into a single
generalized features collection which is written to the database. This
way, enrolled minutiae are more reliable and the fingerprint
recognition quality considerably increases using this enrollment mode.
- Identification ability. As FingerCell is
developed on the VeriFinger basis, it is suitable not only for
fingerprint verification (1-to-1 matching), but also for identification
(1-to-many matching).
- Tolerance to fingerprint translation, rotation and
deformation. Such tolerance is achieved by FingerCell
proprietary fingerprint matching algorithm. The algorithm is able to
identify fingerprints even if they are rotated, translated and
deformed, and matches about 150 fingerprints per second in 1-to-many
mode on a 200 MHz ARM family processor.
- Faster matching using pre-sorted database entries.
For some identification tasks FingerCell effective
matching speed can be increased up to 700 fingerprints per
second (on a 200 MHz ARM family processor) by pre-sorting
database entries using certain global features. Fingerprint matching is
performed first with the database entries having global features most
similar to those of the test fingerprint. If matching within this group
yields no positive result, then the next record with most similar
global features is selected, and so on, until the matching is
successful or the end of the database is reached. In most cases there
is a fairly good chance that the correct match will be found at the
beginning of the search. As a result, the number of comparisons
required to achieve fingerprint identification decreases drastically,
and correspondingly, the matching speed increases.
- Compact portable software. FingerCell is
designed for easy implementation into very various and specific
applications. The algorithm's ANSI C source code is
sensor independent; therefore it can be ported to various platforms and
hardware. Compiled code and internal data arrays require only 400
KB of memory and therefore can be implemented in low memory
microchips, thus reducing hardware costs.
Back to top of this page
|
|
|
|
|
|
|
