Analyzing the potential of active learning for document image classification

被引:4
作者
Saifullah, Saifullah [1 ,2 ]
Agne, Stefan [1 ,3 ]
Dengel, Andreas [1 ,2 ]
Ahmed, Sheraz [1 ,3 ]
机构
[1] German Res Ctr Artificial Intelligence, D-67663 Kaiserslautern, Germany
[2] RPTU Kaiserslautern Landau, D-67663 Kaiserslautern, Germany
[3] DeepReader GmbH, D-67663 Kaiserslautern, Germany
关键词
Document image classification; Document analysis; Active learning; Deep active learning; NEURAL-NETWORKS;
D O I
10.1007/s10032-023-00429-8
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Deep learning has been extensively researched in the field of document analysis and has shown excellent performance across a wide range of document-related tasks. As a result, a great deal of emphasis is now being placed on its practical deployment and integration into modern industrial document processing pipelines. It is well known, however, that deep learning models are data-hungry and often require huge volumes of annotated data in order to achieve competitive performances. And since data annotation is a costly and labor-intensive process, it remains one of the major hurdles to their practical deployment. This study investigates the possibility of using active learning to reduce the costs of data annotation in the context of document image classification, which is one of the core components of modern document processing pipelines. The results of this study demonstrate that by utilizing active learning (AL), deep document classification models can achieve competitive performances to the models trained on fully annotated datasets and, in some cases, even surpass them by annotating only 15-40% of the total training dataset. Furthermore, this study demonstrates that modern AL strategies significantly outperform random querying, and in many cases achieve comparable performance to the models trained on fully annotated datasets even in the presence of practical deployment issues such as data imbalance, and annotation noise, and thus, offer tremendous benefits in real-world deployment of deep document classification models. The code to reproduce our experiments is publicly available at .
引用
收藏
页码:187 / 209
页数:23
相关论文
共 50 条
[41]   Hyperspectral image classification via active learning and broad learning system [J].
Huifang Huang ;
Zhi Liu ;
C. L. Philip Chen ;
Yun Zhang .
Applied Intelligence, 2023, 53 :15683-15694
[42]   Active Learning for Visual Image Classification Method Based on Transfer Learning [J].
Yang, Jihai ;
Li, Shijun ;
Xu, Wenning .
IEEE ACCESS, 2018, 6 :187-198
[43]   Hyperspectral image classification via active learning and broad learning system [J].
Huang, Huifang ;
Liu, Zhi ;
Chen, C. L. Philip ;
Zhang, Yun .
APPLIED INTELLIGENCE, 2023, 53 (12) :15683-15694
[44]   Systemic Risk Document Classification on Indonesian News Articles using Deep Learning and Active Learning [J].
Gumilang, Muhammad ;
Purwarianti, Ayu ;
Nurdinasari, Feriati .
PROCEEDING OF 2019 INTERNATIONAL CONFERENCE ON ELECTRICAL ENGINEERING AND INFORMATICS (ICEEI), 2019, :46-51
[45]   Large-Scale Image Classification Using Active Learning [J].
Alajlan, Naif ;
Pasolli, Edoardo ;
Melgani, Farid ;
Franzoso, Andrea .
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2014, 11 (01) :259-263
[46]   An Active Learning Algorithm for Image Classification Based on Difficulty and Competence [J].
Li, Gen ;
Zhao, Lu ;
Gu, Junwei .
IEEE ACCESS, 2023, 11 :60398-60406
[47]   Joint Posterior Probability Active Learning for Hyperspectral Image Classification [J].
Li, Shuying ;
Wang, Shaowei ;
Li, Qiang .
REMOTE SENSING, 2023, 15 (16)
[48]   HYPERSPECTRAL IMAGE CLASSIFICATION WITH SPARSE REPRESENTATION CLASSIFIER AND ACTIVE LEARNING [J].
Huo, Lian-Zhi ;
Zhao, Li-Jun ;
Tang, Ping .
2016 8TH WORKSHOP ON HYPERSPECTRAL IMAGE AND SIGNAL PROCESSING: EVOLUTION IN REMOTE SENSING (WHISPERS), 2016,
[49]   Multi-criteria active deep learning for image classification [J].
Yuan, Jin ;
Hou, Xingxing ;
Xiao, Yaoqiang ;
Cao, Da ;
Guan, Weili ;
Nie, Liqiang .
KNOWLEDGE-BASED SYSTEMS, 2019, 172 :86-94
[50]   DEEP ADVERSARIAL ACTIVE LEARNING WITH MODEL UNCERTAINTY FOR IMAGE CLASSIFICATION [J].
Zhu, Zheng ;
Wang, Hongxing .
2020 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2020, :1711-1715