Unified deep neural networks for end-to-end recognition of multi-oriented billet identification number

作者：

Highlights：

• Industrial application using image data collected from actual steelworks.

• Network for detecting and recognizing multi-oriented billet identification number.

• Framework for inferring classes and positions of individual characters in text.

摘要

•Industrial application using image data collected from actual steelworks.•Network for detecting and recognizing multi-oriented billet identification number.•Framework for inferring classes and positions of individual characters in text.

论文关键词：Computer vision,Industrial application,Semantic segmentation,Text spotting,End-to-end text recognition

论文评审过程：Received 10 January 2020, Revised 24 November 2020, Accepted 24 November 2020, Available online 6 December 2020, Version of Record 16 December 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114377