Adaptive Biometric Systems Using Ensembles

Autor(es)

Pisani, Paulo Henrique ^[1] ; Lorena, Ana Carolina ^[2] ; de Carvalho, Andre C. P. L. F. ^[3]

Afiliação do(s) autor(es)

^[1] Univ Sao Paulo, Sao Paulo - Brazil

^[2] Univ Fed Sao Paulo, Sao Paulo - Brazil

^[3] Univ Sao Paulo, Dept Comp Sci, Sao Paulo - Brazil

• Tipo de documento: Artigo Científico
• Fonte: IEEE INTELLIGENT SYSTEMS; v. 33, n. 2, p. 19-28, MAR-APR 2018.
• Citações Web of Science: 1

Resumo

With the increased availability of online services, enhanced authentication mechanisms-including biometric systems-are necessary. However, recent studies show that biometric features can change. Consequently, recognition performance can be affected over time. Adaptive biometric systems that can automatically adapt the biometric reference have been proposed to deal with this problem. Frequently, these systems use query samples classified as genuine to adapt the biometric reference. Despite good results, there are concerns regarding their robustness. This article investigates using an ensemble of classifiers to increase these systems' robustness. Ensembles can improve the recognition performance of decision models, providing a more stable classification decision. The authors explore questions regarding the application of ensembles to adaptive biometric systems using one-class classification algorithms, and offer a proposal to automatically adapt the meta classifier over time. (AU)

• Processo FAPESP: 12/22608-8 - Uso de medidas de complexidade de dados no suporte ao aprendizado de máquina supervisionado
• Beneficiário: Ana Carolina Lorena
• Modalidade de apoio: Auxílio à Pesquisa - Jovens Pesquisadores
• Processo FAPESP: 13/07375-0 - CeMEAI - Centro de Ciências Matemáticas Aplicadas à Indústria
• Beneficiário: Francisco Louzada Neto
• Modalidade de apoio: Auxílio à Pesquisa - Centros de Pesquisa, Inovação e Difusão - CEPIDs
• Processo FAPESP: 12/25032-0 - Biometria em um Contexto de Fluxo de Dados com Algoritmos Imunológicos
• Beneficiário: Paulo Henrique Pisani
• Modalidade de apoio: Bolsas no Brasil - Doutorado