作者: Abu Hasnat Mohammad Rubaiyat , Shubhra Aich , Tanjin Taher Toma , Abidur Rahman Mallik , Rafat-Al-Islam
DOI: 10.1109/ICCITECHN.2014.7073086
关键词:
摘要: In this paper, a simple biometric scheme based on RGB retinal fundus images is proposed. First, prominent vasculature energy feature vectors are constructed from to utilize the unique pattern of vasculature. Next, fast normalized cross-correlation matching employed for person identification publicly available DRIVE and STARE databases. This method excels recently published literatures with perfect recognition accuracy 100% 99.77% Its smaller dimension vector better result make eligible real-time scheme.