Automatic recognition of serial numbers in bank notes
作者：
Highlights：
• We study a new topic of automatic recognition of bank note serial numbers.
• A new RMB serial numbers database is released.
• The paper presents an evaluation of state-of-the-art methods on RMB database.
• The enhancements of three types of distortion methods are compared.
• A novel cascade rejection scheme is proposed to achieve high reliability.
摘要
Highlights•We study a new topic of automatic recognition of bank note serial numbers.•A new RMB serial numbers database is released.•The paper presents an evaluation of state-of-the-art methods on RMB database.•The enhancements of three types of distortion methods are compared.•A novel cascade rejection scheme is proposed to achieve high reliability.
论文关键词：Bank note serial number recognition,Cascade rejection,Synthetic training samples,Multiple classifier system
论文评审过程：Received 31 July 2013, Revised 4 January 2014, Accepted 18 February 2014, Available online 28 February 2014.
论文官网地址：https://doi.org/10.1016/j.patcog.2014.02.011
