Content of this page
Fingerprint Recognition Technology for Embedded Systems
Automatic personal identification based on the matching of fingerprints is becoming increasingly popular in systems that control access to physical locations, computer/network resources and bank accounts and systems that register employee attendance time in enterprises. Many of these applications can run on a PC, however some applications require that the system be implemented on low cost, compact and/or mobile embedded devices (doors, gates, cell phones etc.).
Compared to larger, PC-based systems, the fingerprint recognition algorithm for embedded systems requires some specific features and different commonly accepted requirements for things like reliability and speed. Embedded devices usually have weaker processors than personal computers. The fingerprint image processing routines (image enhancement, noise filtration, binarization, skeletonization etc.), used for PC-based applications are quite computationally expensive, therefore require substantial algorithm modification to achieve acceptable image processing time (1 second or less) on the embedded device. This page contains information about the FingerCell embedded fingerprint identification technology, developed on the VeriFinger basis, but having about 4 times faster image processing and feature extraction algorithm.
Why FingerCell?
The FingerCell algorithm, developed on the VeriFinger basis, is designed for embedded biometric systems developers. The algorithm has certain capabilities:
- Reliability. 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 operating curves (ROCs) obtained in testing with two scanner databases compare FingerCell 2.0 (green) and VeriFinger 5.0 (red) reliability under the same conditions.
- Identification ability. As FingerCell is developed on the VeriFinger basis, it is suitable not only for fingerprint verification (1:1 matching), but also for identification (1:N matching). FingerCell can match up to 700 fingerprints per second in 1:N identification mode on 230 MIPS ARM family CPU.
- Image processing speed. Fingerprint image processing time is less than 1 second, which is acceptable for embedded systems.
- Compact software. Compiled code and internal data arrays require only 512 Kb of memory and therefore can be implemented in low memory microchips, thus reducing hardware costs.
- Ready-to-use hardware. FingerCell Device EDK includes a ready-to-use embedded device for customers who don't want to start developing hardware and need a fast and easy implementation of an embedded solution. The stand-alone device includes an integrated U.are.U 4000 sensor.
- Available for various project scales as FingerCell 2.0 Library EDK or FingerCell 2.0 source code EDK.
- Portability. FingerCell Embedded Development Kit is designed for easy implementation into very various and specific applications. The algorithm's source code is written in ANSI C and is sensor independent; therefore it can be ported to various platforms and hardware.
- Embedded and PC-based multi-biometric capable technologies from the same vendor. Combined with our other technologies, FingerCell could be used in developing these advanced systems:
- Mixed embedded/PC systems, using FingerCell EDK together with VeriFinger Standard or Extended SDKs.
- Multi-biometric embedded systems, using FingerCell EDK together with FaceCell EDK.
- Complex multi-biometric embedded/PC systems, using a combination of FingerCell EDK, FaceCell EDK, VeriFinger SDK and VeriLook SDK .
Algorithm
The FingerCell algorithm is similar to the VeriFinger algorithm and includes these features:
- FingerCell is fully tolerant to fingerprint translation, rotation and deformation. Such tolerance is achieved by our proprietary fingerprint matching algorithm.
- FingerCell does not require the presence of fingerprint core or delta points in the image and can recognize a fingerprint from any part of it.
- FingerCell has fingerprint enrollment with features generalization mode. This 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.
- FingerCell can use database entries which were pre-sorted 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 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 effective matching speed increases.
- The FingerCell embedded algorithm is similar to VeriFinger , but it has about 4 times faster image processing and feature extraction algorithms.
Specifications
Please note, that these specifications were determined on hardware with 230 MIPS ARM family processor.
| Enrollment time |
< 1 second |
| Enrollment time in features generalization mode |
< 3 seconds |
| Verification time |
< 1 second |
| Matching speed |
up to 700 fingerprints/second |
| Template size |
300 - 600 bytes |
FingerCell Trial
Neurotechnologija offers a FingerCell 2.0 EDK trial. This 30 day trial kit allows developers to explore the technology and to try it in real environments and real applications.
FingerCell 2.0 EDK trial is available for downloading.
Related Products
These products are based on the FingerCell 2.0 algorithm:
|
