Bangladesh Voter Registration Project using Biometrics to Detect and Prevent Duplicate Registrations

The Bangladesh Army has selected MegaMatcher SDK multi-biometric technology to identify duplicate registrations in the nation's voter database. The Bangladesh Voter Registration Project registered more than 80 million citizens using biometric face and fingerprint technology. After evaluating a number of biometric technologies for their duplicate search system, the Bangladesh Army determined that MegaMatcher from Neurotechnology was able to identify more duplicate registrations with a higher degree of accuracy than any other system tested. System integrator Dohatec New Media was hired to help design and implement the MegaMatcher-based system. To date, more than 48 million voter registration records have been matched.

"Matching performed using MegaMatcher SDK could find more valid duplicate entries than any other SDK used for the same purpose," said Lt. Col. Md Mostafizur Rahman, staff officer grade 1 (Information Technology Directorate) for the Bangladesh Army. "It has enhanced the technological capabilities of our system and the cost-per-unit for the SDK is cost- effective."

The Dohatec Biometrics Fusion Server system uses MegaMatcher Client to generate templates from face and fingerprint images that were captured with a BIO-Key system, then the match technology is used to search the database and identify duplicate records. Bangladesh runs MegaMatcher on Microsoft Windows XP and Microsoft Windows Server with Microsoft SQL Server as the back-end database.

MegaMatcher provides the high speed and reliability required for the development of national-scale automated fingerprint identification systems (AFIS) and multi-biometric face/fingerprint identification systems. Suitable for both civil and forensic use, the system includes both fingerprint and face identification engines with a fusion algorithm that allows the two technologies to work together to provide very fast 1:N (1 to many) matching with even higher reliability than AFIS or facial recognition alone.