Document is current
Any future updates will be listed below
-
Authors
Wang, Xinxin
Chen, Liang
Liu, Changhong
Liu, Jinyu
-
Funding
Funding for this research was provided by:
This work was supported by the National Natural Science Foundation of China (51675108)
Key Scientific Research Program of the Education Department of Shaanxi Province,China (22JS021)
- License Information
-
More Information
Article History
Received: 11 February 2024
Accepted: 16 December 2024
First Online: 6 January 2025
Declarations
:
: The authors declare no competing interests.
: The issue of inaccurate recognition of curved text in novel charts has been effectively addressed through the application of character-level text annotation methods.The visual dashboard underwent segmentation and organization using an object detection algorithm. This step was taken to provide chart support for subsequent multi-table cross-referencing. The enhancement of mathematical computational capabilities in chart-based question and answer processes was achieved by introducing an inference fusion head. This additional component significantly contributes to the system’s ability to perform intelligent analysis and respond to queries.
