HUST-OBS
收藏资源简介:
HUST-OBS数据集由华中科技大学创建,包含140,053张甲骨文图像,分为已解读和未解读两部分,旨在通过人工智能辅助解读甲骨文。数据集内容丰富,涵盖1,588个已解读字符和9,411个未解读字符,图像来源于书籍、网站和数据库。创建过程中,采用了半自动数据收集和标注流程,确保数据质量和多样性。该数据集主要应用于甲骨文识别和解读研究,旨在解决甲骨文解读中的复杂性和挑战。
The HUST-OBS dataset was developed by Huazhong University of Science and Technology. It contains 140,053 oracle bone inscription (OBI) images, which are categorized into two groups: interpreted and uninterpreted. The dataset is designed to facilitate oracle bone inscription interpretation with the assistance of artificial intelligence. The dataset covers rich content, including 1,588 interpreted characters and 9,411 uninterpreted characters. Its source materials are derived from books, websites and databases. During the construction process, a semi-automatic data collection and annotation workflow was employed to guarantee data quality and diversity. This dataset is primarily utilized for research on oracle bone inscription recognition and interpretation, with the goal of tackling the complexity and challenges inherent in oracle bone inscription interpretation.




