MultiSHTM: Multi-Level Attention Enabled Bi-Directional Model for the Summarization of Chart Images
Chart-to-text conversion is an emerging research area focused on extracting useful information from chart images to improve understanding and analysis. Deep learning methods help in identifying important details and patterns from charts. However, existing models struggle to analyze charts because of...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11000319/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850100979941244928 |
|---|---|
| author | Indra Kumari Hansung Lee |
| author_facet | Indra Kumari Hansung Lee |
| author_sort | Indra Kumari |
| collection | DOAJ |
| description | Chart-to-text conversion is an emerging research area focused on extracting useful information from chart images to improve understanding and analysis. Deep learning methods help in identifying important details and patterns from charts. However, existing models struggle to analyze charts because of the mix of text and graphical elements. To solve this problem, we propose a new method called MultiSHTM (Multi-level Stacked Houghless Network-based Bi-LSTM). This method improves accuracy, reduces complexity, and works well across different chart types. MultiSHTM integrates two key innovations: 1) a multilevel attention mechanism in a stacked houghless network, which accurately identifies key points in charts without relying on traditional Hough Transform-based methods and 2) a Bi-LSTM model enhanced with a Hierarchical and Channel Attention module, which effectively captures contextual relationships to generate precise summaries of chart images. Compared to existing methods, MultiSHTM performs better, achieving scores of Rouge: 0.55, Bleu: 0.45, Cider: 0.8, Meteor: 0.25, and Spice: 25.60. |
| format | Article |
| id | doaj-art-433f77945d314fbfa9eeb888515798ca |
| institution | DOAJ |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-433f77945d314fbfa9eeb888515798ca2025-08-20T02:40:10ZengIEEEIEEE Access2169-35362025-01-011310178910180010.1109/ACCESS.2025.356884911000319MultiSHTM: Multi-Level Attention Enabled Bi-Directional Model for the Summarization of Chart ImagesIndra Kumari0https://orcid.org/0000-0001-5418-0981Hansung Lee1https://orcid.org/0000-0002-6519-4120Department of Artificial Intelligence Engineering, Sunchon National University (SCNU), Suncheon, Republic of KoreaDepartment of Computer Education, Gyeongkuk National University, Andong, Gyeongsangbuk-do, Republic of KoreaChart-to-text conversion is an emerging research area focused on extracting useful information from chart images to improve understanding and analysis. Deep learning methods help in identifying important details and patterns from charts. However, existing models struggle to analyze charts because of the mix of text and graphical elements. To solve this problem, we propose a new method called MultiSHTM (Multi-level Stacked Houghless Network-based Bi-LSTM). This method improves accuracy, reduces complexity, and works well across different chart types. MultiSHTM integrates two key innovations: 1) a multilevel attention mechanism in a stacked houghless network, which accurately identifies key points in charts without relying on traditional Hough Transform-based methods and 2) a Bi-LSTM model enhanced with a Hierarchical and Channel Attention module, which effectively captures contextual relationships to generate precise summaries of chart images. Compared to existing methods, MultiSHTM performs better, achieving scores of Rouge: 0.55, Bleu: 0.45, Cider: 0.8, Meteor: 0.25, and Spice: 25.60.https://ieeexplore.ieee.org/document/11000319/Summarizationkey point detectionspatial attentionhierarchical attentiondeep learning |
| spellingShingle | Indra Kumari Hansung Lee MultiSHTM: Multi-Level Attention Enabled Bi-Directional Model for the Summarization of Chart Images IEEE Access Summarization key point detection spatial attention hierarchical attention deep learning |
| title | MultiSHTM: Multi-Level Attention Enabled Bi-Directional Model for the Summarization of Chart Images |
| title_full | MultiSHTM: Multi-Level Attention Enabled Bi-Directional Model for the Summarization of Chart Images |
| title_fullStr | MultiSHTM: Multi-Level Attention Enabled Bi-Directional Model for the Summarization of Chart Images |
| title_full_unstemmed | MultiSHTM: Multi-Level Attention Enabled Bi-Directional Model for the Summarization of Chart Images |
| title_short | MultiSHTM: Multi-Level Attention Enabled Bi-Directional Model for the Summarization of Chart Images |
| title_sort | multishtm multi level attention enabled bi directional model for the summarization of chart images |
| topic | Summarization key point detection spatial attention hierarchical attention deep learning |
| url | https://ieeexplore.ieee.org/document/11000319/ |
| work_keys_str_mv | AT indrakumari multishtmmultilevelattentionenabledbidirectionalmodelforthesummarizationofchartimages AT hansunglee multishtmmultilevelattentionenabledbidirectionalmodelforthesummarizationofchartimages |