CT-Net: Cascade T-shape deep fusion networks for document binarization

作者：

Highlights：

• We present a Cascade T-Shape Deep Networks framework for document image binarization.

• The T-Net is multi-task framework for document enhancement and binarization. Deep features learned by the enhancement are transferred to the binarization which can improve the performance of binarization.

• The Cascade T-Net is evaluated on nine public DIBCO datasets and achieves new state-of-the-art performance.

摘要

•We present a Cascade T-Shape Deep Networks framework for document image binarization.•The T-Net is multi-task framework for document enhancement and binarization. Deep features learned by the enhancement are transferred to the binarization which can improve the performance of binarization.•The Cascade T-Net is evaluated on nine public DIBCO datasets and achieves new state-of-the-art performance.

论文关键词：Cascade T-Net,Binarization,Enhancement,DIBCO,Convolutional neural networks

论文评审过程：Received 14 June 2020, Revised 19 October 2020, Accepted 27 April 2021, Available online 5 May 2021, Version of Record 26 May 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.108010