In the developing world, namely Sub-Saharan Africa (SSA), the lack of official identification promotes a cycle of poverty and social exclusion for many individuals. Robust national identification systems are necessary for social, political, health and economic development, as well as aid delivery. Without a unique identification, such as the hundreds of millions of poor people who lack birth certificates in SSA, summed up with no addresses and single-word names, many individuals cannot exercise basic rights nor access several formal services, from health to financial insurance, formal employment, or democratic participation.
Biometric identification is considerably more accurate and secure than traditional methods of individual identification and authentication. When combined with technology such as mobile phones, biometrics can help streamline and extend services to remote, under-served locations, reaching marginalized groups, reducing fraud and corruption, and improving security. Biometric features can be extracted, for example, from iris scan, fingerprints, facial recognition, voice analysis, hand geometry, gait analysis, etc. One of the difficulties of biometric identification systems lies in the identification of children, elderly or disabled people. Additionally, children identification is very necessary in SSA for health and education initiatives, such as vaccination procedures.
The fusion of multiple biometrics, multimodal biometrics, helps to minimize identification error rates. Fusion methods include combining scores and classifiers or processing biometric modalities sequentially until an acceptable match is obtained.
Author: Rui Esteves
Type: MSc thesis
Partner: Faculdade de Engenharia da Universidade do Porto