摘要
随着图书馆信息化的快速发展,高校图书馆的安防和现代化管理是智慧图书馆发展的一个方向。在人工智能时代下,图书馆刷脸趋势越来越明显。高校图书馆的人脸识别场景与传统人脸识别场景有所不同,可结合图书馆场景对人脸识别技术进行改进。本文结合高校学生出入馆时间规律和行为习惯等的特点,利用Redis集群技术,搭建时空缓存数据库,动态调整人脸序列缓存,进而减少对人脸识别的响应时间。经实验测试,在不改变人脸识别算法的前提下,通过工程化的改良,可以节省服务时间,提升服务效率。
With the rapid development of Library informationization,the security and modern management of university libraries is a direction of the development of intelligent libraries.In the era of artificial intelligence,the trend of Library brushing is more and more obvious.Face recognition scenes in university libraries are different from those in traditional ones of should be improved.Based on the characteristics of the time rules and behavior habits of University students,this paper uses Redis cluster technology to build a spatiotemporal cache database,dynamically adjust the face sequence cache,and then reduce the response time to face recognition.After experimental testing,without changing the face recognition algorithm,through engineering improvement,the method can save service time and improve service efficiency.
出处
《信息技术与信息化》
2019年第9期46-48,共3页
Information Technology and Informatization
基金
教育部产学合作协同育人项目(201802048012)
江苏省独立学院民办高校2019图书馆项目(LIB-201909)课题资助