FingerCell SDK

Fingerprint identification for embedded platforms

FingerCell technology is designed for embedded biometric systems developers and features compact, sensor-independent and cross-platform fingerprint recognition algorithm. It offers decent performance on various embedded devices based on low-power microcontrollers or processors.

FingerCell is available for integrators as Software Development Kits (SDK) with FingerCell library or source code for developing a fast and reliable system on embedded or mobile platform.

FingerCell Demo Unit

Neurotechnology offers FingerCell Demo Unit – testing hardware with pre-installed FingerCell algorithm for the technology evaluation. Please contact us for more information about obtaining the Demo Unit.

General view FingerCell 3.1 demo unit, general view
Click to zoom


Top-down view FingerCell 3.1 demo unit, top down view
Click to zoom

FingerCell 3.1 demo unit, general view

FingerCell Demo Unit is based on an STM Discovery evaluation board with these components:

  • STM32F407 microcontroller running at 168 MHz and having 192 kB of RAM;
  • Next Biometrics fingerprint sensor connected via SPI, which produces 180 x 256 pixels grayscale image at 385 ppi resolution;
  • 2x16 LCD screen;
  • 4 push buttons.

The FingerCell algorithm in the Demo Unit performs single fingerprint template extraction in 700 milliseconds and matches 250 fingerprints per second.

The Demo Unit can be used in two modes:

  • Connected to a host PC – all operations, including template import and export, are performed via USB communication.
    • PC-side API and a programming sample, which shows how to use all commands, are provided.
    • The host PC needs to run Linux OS.
  • Stand-alone – these actions are performed by the unit after pressing the corresponding buttons:
    • Enroll – performs single fingerprint capture and enrollment. The LCD screen indicates that enrollment is in progress and asks to put a finger on scanner. Once the finger is captured, the template will be extracted, duplicates will be checked in the database and the finger will be enrolled if no duplicates are found. An auto-generated ID for the enrolled fingerprint will be displayed.
    • Enroll-stitch – performs multiple capture of the same fingerprint, template stitching and enrollment. A user should scan his/her finger 8 times. The LCD screen will show the enrollment progress, and a template will be generated for each scanned fingerprint. After 8 scans, all templates will be merged into single generalized template, which will be enrolled into database and the auto-generated ID will be displayed.
    • Identify – performs single fingerprint capture and identification against the templates stored in the database. A user should scan a fingerprint, which will be processed and identified against the previously enrolled fingerprints. The LCD screen will show the identification process progress. If match a is found, the matching fingerprint ID and matching score will be displayed. The higher matching score is, the more similar the fingerprints are.
    • Cancel/clear:
      • Single click – cancel the selected operation.
      • Double click – delete all entries from the database.