Information extraction from historical handwritten document images with a context-aware neural model

作者：

Highlights：

• Two methods to extract information from structured handwritten document images.

• They are lexicon free, work at image level without any intermediate transcription.

• They model the structure of information elements by analyzing the words relations.

• Validation using a public dataset, outperforming the state of the art.

摘要

•Two methods to extract information from structured handwritten document images.•They are lexicon free, work at image level without any intermediate transcription.•They model the structure of information elements by analyzing the words relations.•Validation using a public dataset, outperforming the state of the art.

论文关键词：Document image analysis,Handwritten documents,Named entity recognition,Deep neural networks

论文评审过程：Received 8 January 2018, Revised 24 June 2018, Accepted 27 August 2018, Available online 31 August 2018, Version of Record 7 September 2018.

论文官网地址：https://doi.org/10.1016/j.patcog.2018.08.020