摘要
指纹自动识别技术在身份认证领域有广泛和重要的应用价值.文中在综述该领域发展现状的基础上,重点研究了指纹图像识别的算法评价问题.提出了全面衡量指纹识别性能的必要性和方法,其中包括指纹数据库的建立要求,以及拒真率、认假率和拒登率等9项必要的评价指标,并用实例说明了这种评价体系的使用过程.
The automatic fingerprint recognition techniques are widely used in the field of personal identification. Based on the summarization of the recent developments of the field, this paper discusses the performance evalution of fingerprint image recognition algorithms in detail. The necessities and the methodologies for evaluating the performance of an algorithm in a comprehensive manner are presented, which include the way to build a fingerprint database and nine necessary indices such as False Reject Rate, False Accept Rate, False Enroll Rate, etc. Finally the whole process of doing so is illustrated with an example.
出处
《西安电子科技大学学报》
EI
CAS
CSCD
北大核心
2002年第6期804-808,共5页
Journal of Xidian University
基金
国家自然科学基金资助项目(69982007)
国家自然科学基金重大项目(69896240)
国防跨行业基金资助项目
关键词
指纹识别系统
性能评价
图像处理
人工智能
图像识别
fingerprint recognition
image processing and artificial intelligence
performance evaluation
