作者: P. A. Johnson , F. Hua , S. Schuckers
DOI: 10.1109/IJCB.2011.6117481
关键词:
摘要: Multimodal systems have been used for the increased robustness of biometric recognition tasks. A unique strength multimodal can be found when presented with samples degraded quality in a subset modalities. This study looks at effect degradation on system performance using Q-FIRE database. The database is composed face and iris biometrics captured defined levels, controlled acquisition. allows assessment pertaining to image factors. Methods measuring based illumination conditions are explored as well strategies incorporating these metrics into fusion algorithm. paper provides further evidence dataset that utilizing sample scheme improves non-ideal acquisition environments.