Enhance to read better: A Multi-Task Adversarial Network for Handwritten Document Image Enhancement

作者：

Highlights：

• A Generative Adversarial Network for handwritten document image binarization.

• We perform document binarization while ensuring text readability, simultaneously, by integrating a handwritten text recognition component within the proposed architecture.

• The proposed model enhances different forms of documents, independently of the text language.

• We achieve state-of-the-art performance on the public H-DIBCO datasets.

摘要

•A Generative Adversarial Network for handwritten document image binarization.•We perform document binarization while ensuring text readability, simultaneously, by integrating a handwritten text recognition component within the proposed architecture.•The proposed model enhances different forms of documents, independently of the text language.•We achieve state-of-the-art performance on the public H-DIBCO datasets.

论文关键词：Handwritten document image binarization,Document enhancement,Handwriting text recognition,Generative adversarial networks,Recurrent neural networks

论文评审过程：Received 13 May 2021, Revised 13 August 2021, Accepted 10 October 2021, Available online 11 October 2021, Version of Record 17 October 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.108370