dhSegment: A generic deep-learning approach for document segmentation

被引:123
|
作者
Oliveira, Sofia Ares [1 ]
Seguin, Benoit [1 ]
Kaplan, Frederic [1 ]
机构
[1] Ecole Polytech Fed Lausanne, Digital Humanities Lab, Lausanne, VD, Switzerland
关键词
document segmentation; historical document processing; document layout analysis; neural network; deep learning;
D O I
10.1109/ICFHR-2018.2018.00011
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In recent years there have been multiple successful attempts tackling document processing problems separately by designing task specific hand-tuned strategies. We argue that the diversity of historical document processing tasks prohibits to solve them one at a time and shows a need for designing generic approaches in order to handle the variability of historical series. In this paper, we address multiple tasks simultaneously such as page extraction, baseline extraction, layout analysis or multiple typologies of illustrations and photograph extraction. We propose an open-source implementation of a CNN-based pixel-wise predictor coupled with task dependent post-processing blocks. We show that a single CNN-architecture can be used across tasks with competitive results. Moreover most of the task-specific post-precessing steps can be decomposed in a small number of simple and standard reusable operations, adding to the flexibility of our approach.
引用
收藏
页码:7 / 12
页数:6
相关论文
共 50 条
  • [41] Deep-learning approach in the study of skin lesions
    Filipescu, Stefan-Gabriel
    Butacu, Alexandra-Irina
    Tiplica, George-Sorin
    Nastac, Dumitru-Iulian
    SKIN RESEARCH AND TECHNOLOGY, 2021, 27 (05) : 931 - 939
  • [42] A new multimodal deep-learning model to video scene segmentation
    Trojahn, Tiago H.
    Kishi, Rodrigo M.
    Goularte, Rudinei
    WEBMEDIA'18: PROCEEDINGS OF THE 24TH BRAZILIAN SYMPOSIUM ON MULTIMEDIA AND THE WEB, 2018, : 205 - 212
  • [43] Deep-learning based automated segmentation of Diabetic Retinopathy symptoms
    Yeh, Hung
    Lin, Cheng-Jhong
    Hsu, Chih-Chung
    Lee, Chia-Yen
    2020 INTERNATIONAL SYMPOSIUM ON COMPUTER, CONSUMER AND CONTROL (IS3C 2020), 2021, : 497 - 499
  • [44] Portrait Segmentation Using Ensemble of Heterogeneous Deep-Learning Models
    Kim, Yong-Woon
    Byun, Yung-Cheol
    Krishna, Addapalli V. N.
    ENTROPY, 2021, 23 (02) : 1 - 20
  • [45] Open-source deep-learning software for bioimage segmentation
    Lucas, Alice M.
    Ryder, Pearl, V
    Li, Bin
    Cimini, Beth A.
    Eliceiri, Kevin W.
    Carpenter, Anne E.
    MOLECULAR BIOLOGY OF THE CELL, 2021, 32 (09) : 823 - 829
  • [46] A hybrid deep segmentation network for fundus vessels via deep-learning framework
    Yang, Lei
    Wang, Huaixin
    Zeng, Qingshan
    Liu, Yanhong
    Bian, Guibin
    NEUROCOMPUTING, 2021, 448 : 168 - 178
  • [47] A combined deep-learning approach to fully automatic left ventricle segmentation in cardiac magnetic resonance imaging
    Moreno, Ramon A.
    de Sa Rebelo, Marina F. S.
    Carvalho, Talles
    Assuncao-Jr, Antonildes N.
    Dantas Jr, Roberto N.
    do Val, Renata
    Marin, Angela S.
    Bordignom, Adriano
    Nomura, Cesar H.
    Gutierrez, Marco A.
    MEDICAL IMAGING 2019: BIOMEDICAL APPLICATIONS IN MOLECULAR, STRUCTURAL, AND FUNCTIONAL IMAGING, 2019, 10953
  • [48] Advancing Barrett's Esophagus Segmentation: A Deep-Learning Ensemble Approach with Data Augmentation and Model Collaboration
    Lee, Jiann-Der
    Tsai, Chih Mao
    BIOENGINEERING-BASEL, 2024, 11 (01):
  • [49] A Deep-Learning Model with Learnable Group Convolution and Deep Supervision for Brain Tumor Segmentation
    Liu, Hengxin
    Li, Qiang
    Wang, I-Chi
    MATHEMATICAL PROBLEMS IN ENGINEERING, 2021, 2021
  • [50] Evaluation of a Deep-Learning Auto-Segmentation Model of Cardiac Substructures
    Tam, A.
    Liu, J. R.
    Ketcherside, T.
    Eustace, N. J.
    Chen, Q.
    Chen, Y. J.
    Liu, A.
    INTERNATIONAL JOURNAL OF RADIATION ONCOLOGY BIOLOGY PHYSICS, 2023, 117 (02): : E724 - E725