A Recurrent Neural Network based deep learning model for offline signature verification and recognition system

作者：

Highlights：

• A deep learning based model to recognize and verify offline signatures is proposed.

• Recurrent Neural Network is used to achieve the task.

• Structure and direction oriented features have been extracted.

• The performance has been tested on six widely used public signature databases.

摘要

•A deep learning based model to recognize and verify offline signatures is proposed.•Recurrent Neural Network is used to achieve the task.•Structure and direction oriented features have been extracted.•The performance has been tested on six widely used public signature databases.

论文关键词：Offline signature,Verification and recognition,Deep learning,RNN,LSTM and BLSTM

论文评审过程：Received 27 May 2020, Revised 16 October 2020, Accepted 4 November 2020, Available online 7 November 2020, Version of Record 24 January 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114249