NetLay: Layout Classification Dataset for Enhancing Layout Analysis
Within the domain of historical document image analysis, the process of identifying the spatial structure of a document image is an essential step in many document processing tasks, such as optical character recognition and information extraction. Advancements in layout analysis promise to enhanc...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Fondazione Università Ca’ Foscari
2024-12-01
|
| Series: | magazén |
| Subjects: | |
| Online Access: | http://doi.org/10.30687/mag/2724-3923/2024/02/003 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846118639502098432 |
|---|---|
| author | Gogawale, Sharva Bambaci, Luigi Kurar-Barakat, Berat Vasyutinsky Shapira, Daria Stökl Ben Ezra, Daniel Dershowitz, Nachum |
| author_facet | Gogawale, Sharva Bambaci, Luigi Kurar-Barakat, Berat Vasyutinsky Shapira, Daria Stökl Ben Ezra, Daniel Dershowitz, Nachum |
| author_sort | Gogawale, Sharva |
| collection | DOAJ |
| description |
Within the domain of historical document image analysis, the process of identifying the spatial structure of a document image is an essential step in many document processing tasks, such as optical character recognition and information extraction. Advancements in layout analysis promise to enhance efficiency and accuracy using specialized models tailored to distinct layouts. We introduce NetLay, a new dataset for benchmarking layout classification algorithms for historical works. It consists of over 1,300 images of pages of printed Hebrew (or Hebrew‑character) books in a variety of styles, categorized into four different classes based on their layout (the number of text columns and regions). Ground truth was crafted manually at the page level. Furthermore, we conduct an in‑depth performance evaluation of various layout classification algorithms, which are based on deep‑learning models that learn to extract spatial features from images. We evaluate our algorithms on NetLay and achieve state‑of‑the‑art results on the task of layout classification for historical books.
|
| format | Article |
| id | doaj-art-5a14460f4446400eb14170cb6a990d3d |
| institution | Kabale University |
| issn | 2724-3923 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Fondazione Università Ca’ Foscari |
| record_format | Article |
| series | magazén |
| spelling | doaj-art-5a14460f4446400eb14170cb6a990d3d2024-12-17T12:17:27ZengFondazione Università Ca’ Foscarimagazén2724-39232024-12-015210.30687/mag/2724-3923/2024/02/003journal_article_18143NetLay: Layout Classification Dataset for Enhancing Layout AnalysisGogawale, Sharva0Bambaci, Luigi1Kurar-Barakat, Berat2Vasyutinsky Shapira, Daria3Stökl Ben Ezra, Daniel4Dershowitz, Nachum5Tel Aviv University, IsraelÉcole Pratique des Hautes Études (EPHE), FranceTel Aviv University, IsraelTel Aviv University, IsraelÉcole Pratique des Hautes Études (EPHE), FranceTel Aviv University, Israel Within the domain of historical document image analysis, the process of identifying the spatial structure of a document image is an essential step in many document processing tasks, such as optical character recognition and information extraction. Advancements in layout analysis promise to enhance efficiency and accuracy using specialized models tailored to distinct layouts. We introduce NetLay, a new dataset for benchmarking layout classification algorithms for historical works. It consists of over 1,300 images of pages of printed Hebrew (or Hebrew‑character) books in a variety of styles, categorized into four different classes based on their layout (the number of text columns and regions). Ground truth was crafted manually at the page level. Furthermore, we conduct an in‑depth performance evaluation of various layout classification algorithms, which are based on deep‑learning models that learn to extract spatial features from images. We evaluate our algorithms on NetLay and achieve state‑of‑the‑art results on the task of layout classification for historical books. http://doi.org/10.30687/mag/2724-3923/2024/02/003Convolutional neural networks. Deep learning. Historical document analysis. Layout analysis. Layout classification. Multi‑label classification |
| spellingShingle | Gogawale, Sharva Bambaci, Luigi Kurar-Barakat, Berat Vasyutinsky Shapira, Daria Stökl Ben Ezra, Daniel Dershowitz, Nachum NetLay: Layout Classification Dataset for Enhancing Layout Analysis magazén Convolutional neural networks. Deep learning. Historical document analysis. Layout analysis. Layout classification. Multi‑label classification |
| title | NetLay: Layout Classification Dataset for Enhancing Layout Analysis |
| title_full | NetLay: Layout Classification Dataset for Enhancing Layout Analysis |
| title_fullStr | NetLay: Layout Classification Dataset for Enhancing Layout Analysis |
| title_full_unstemmed | NetLay: Layout Classification Dataset for Enhancing Layout Analysis |
| title_short | NetLay: Layout Classification Dataset for Enhancing Layout Analysis |
| title_sort | netlay layout classification dataset for enhancing layout analysis |
| topic | Convolutional neural networks. Deep learning. Historical document analysis. Layout analysis. Layout classification. Multi‑label classification |
| url | http://doi.org/10.30687/mag/2724-3923/2024/02/003 |
| work_keys_str_mv | AT gogawalesharva netlaylayoutclassificationdatasetforenhancinglayoutanalysis AT bambaciluigi netlaylayoutclassificationdatasetforenhancinglayoutanalysis AT kurarbarakatberat netlaylayoutclassificationdatasetforenhancinglayoutanalysis AT vasyutinskyshapiradaria netlaylayoutclassificationdatasetforenhancinglayoutanalysis AT stoklbenezradaniel netlaylayoutclassificationdatasetforenhancinglayoutanalysis AT dershowitznachum netlaylayoutclassificationdatasetforenhancinglayoutanalysis |