Character confidence based on N-best list for keyword spotting in online Chinese handwritten documents

作者：

Highlights：

• The posterior probabilities of candidate characters are computed based on the N-best list.

• The confidence parameter is estimated from the character confusion network.

• The proposed method outperforms those based on word confidence and text line recognition.

摘要

Highlights•The posterior probabilities of candidate characters are computed based on the N-best list.•The confidence parameter is estimated from the character confusion network.•The proposed method outperforms those based on word confidence and text line recognition.

论文关键词：Online Chinese handwritten documents,Keyword spotting,Posterior probability,N-best list,Confidence measure,Confusion network

论文评审过程：Received 22 January 2013, Revised 11 November 2013, Accepted 2 December 2013, Available online 11 December 2013.

论文官网地址：https://doi.org/10.1016/j.patcog.2013.12.001