2.ZK Finger SDK allows easy integration to the applications of the users. It opens the interface of dealing with various image, it can support scanner and fingerprint sensor (The resolution of the image>= 300DPI).

3.ZK Finger SDK algorithm is able to filter noises, enhance the contrast of ridge and trough. Even from the fingerprint with bad quality (dirty, cuts, scars, dry, wet), the algorithm is able to extract accurate minutia points.

4.ZK Finger SDK Matching algorithm allows the identification or verification of 0 to 360 degree rotation of the live fingerprint. With state-of-the-art techniques, the matching can still be conducted at a high speed despite the different orientation angles of the finger (Approx Matching Speed 3000 Match/sec, even for fingers with low minutia points (<=10), a fingerprint normally would have >= 15 minutia points.)

5.ZK Finger SDK Matching algorithm does not require complete minutia points such as (core, delta, etc); it can make identification successfully based on several essential or critical minutia points.

6.ZK Finger SDK categorization method (Fingerprints are divided mainly into 5 categories: (Arch, left loop, right loop, tented arch, whorl), if categorization is used in the first instance, the matching speed will be greatly increased.

7.ZK Finger SDK Matching algorithm source code is simple, only 200Kbytes of program space is required; therefore it is easily accommodated into any PC hardware and software programs.

We obtained 2000 fingerprints from 4 types of Sensor (YLC, DFR200, U.ARE.U, Authentec) to test the performance of the ZK Finger SDK Matching algorithm (For Each Sensor, we obtained 500 fingerprints). We make each fingerprint in the database cross matching with the 2000 fingerprints. About 4,000,000 matches were executed; the following are the results:

Fingerprint template size 256 or 1024 bytes
Rotation 0 - 360 degrees
FAR <= 0.001%
FRR <= 2.0%
Enrollment time 0.5 Seconds
Matching Speed 2000-4500 Matches/sec
Livescan Image > = 300DPI