Leveraging the Cloud for Big Data Biometrics: Meeting the Performance Requirements of the Next Generation Biometric Systems

摘要

This paper presents a prototype system for generalized searching of cloud-scale biometric data as well as an application of this system to the task of matching a collection of synthetic human iris images. First, we discuss growth trends in the biometric technology market that warrant the application of new data storage and retrieval technologies. Next, we discuss the considerations that motivated our decisions when designing our prototype framework. We then discuss the underlying open source components that we selected for the prototype, its API design, and the relationship between the API and the underlying architecture. We provide details of our implementation of human iris matching within our prototype framework. Finally, we discuss opportunities for future research.