基于多阶段数据生成的自循环文本智能识别  

Self-circulation Intelligent Text Recognition Based on Multi-stage Data Generation

作　　者：马新强[1,2,3] 刘丽娜 李雪维 顾晔 黄羿 刘勇[2] MA Xinqiang;LIU Lina;LI Xuewei;GU Ye;HUANG Yi;LIU Yong(College of Computer Science and Technology,Guizhou University,Guiyang 550025;Institute of Cyber-Systems and Control,Zhejiang University,Hangzhou 310027;Institute of Intelligent Computing and Visualization Based on Big Data,Chongqing University of Arts and Sciences,Chong-qing 402160;Material Branch,State Grid Zhejiang Electric Power Co.Ltd.,Hangzhou 310000)

机构地区：[1]贵州大学计算机科学与技术学院,贵阳550025 [2]浙江大学智能系统与控制研究所,杭州310027 [3]重庆文理学院大数据智能计算与可视化研究所,重庆402160 [4]国网浙江省电力有限公司物资分公司,杭州310000

出　　处：《模式识别与人工智能》2020年第5期468-477,共10页Pattern Recognition and Artificial Intelligence

基　　金：浙江省重点研发计划项目(No.2019C01004);广东省重点研发计划项目(No.2019B010120001);重庆市发改委重大产业技术研发项目(No.2018148208);重庆市技术创新与应用发展重点项目(No.cstc2019jscx-fxydX0094);重庆英才创新创业示范团队(No.CQYC201903167);浙江大学工业控制技术国家重点实验室开放课题(No.ICT170330,ICT1800413,ICT1900358)资助。

摘　　要：在复杂多样场景下,极少存在同时对英文和中文都具有较优识别效果的大数据标注方法.因此文中提出针对复杂多样文本识别场景的数据生成和多阶段自循环训练算法.按照定义的生成数据参数随机生成文本数据,免去数据标注过程.在卷积循环神经网络的基础上,进行多阶段自循环训练,在循环过程中通过控制数据生成策略不断提升样本的识别精度.实验表明,文中算法在多个公开英文数据集及中文特定的复杂文本场景下都具有良好的识别性能.There are few effective big data annotation methods for both English and Chinese recognition in complex and diverse scenarios.Therefore,multi-stage data generation self-circulation training algorithm(MSDG-OCR)for complex and diverse text recognition scenarios is proposed.Text data is generated randomly according to the defined generated data parameters,and the data annotation process is omitted.Grounded on convolutional recurrent neural network(CRNN)model,multi-stage self-circulation training is carried out,and the recognition accuracy of the samples is continuously improved by controlling the data generation strategy during the loop process.Experiments show that the proposed algorithm gains good recognition performance in multiple public English datasets and Chinese-specific complex text scenarios.

关 键 词：大数据标注 数据生成 文本识别 卷积循环神经网络(CRNN) 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]