BullingerDB
收藏资源简介:
BullingerDB是由弗里堡大学等机构联合创建的大规模历史文档分析基准数据集,基于16世纪瑞士改革家海因里希·布林格的信件手稿。该数据集包含20,898页手稿图像,涵盖499,222个文本行,由796位书写者在六十年间用拉丁语和早期新高地德语等多种语言撰写,并附有书写者身份和时间等元数据。数据集的创建过程包括对数字化扫描件进行布局分析、文本行分割以及通过学者转录与自动对齐相结合的方式生成标注,最终经过严格的后处理筛选。该数据集主要应用于手写文本识别和书写者检索研究领域,旨在解决历史手稿的多语言识别、长期笔迹风格演变分析等挑战,为文档分析提供具有丰富时空维度的基准资源。
This dataset was jointly constructed by the University of Paris Cité and the University of Groningen. It contains precomputed optical parameter grids for seven key aerosol species in exoplanetary atmospheres, such as amorphous silicates, titanium tholins and other similar species. The data covers a wavelength range of 0.3–50 micrometers and a particle size range of 0.001–30 micrometers. Utilizing a high-precision interpolation method, the dataset achieves a 1.4–17-fold increase in computational speed. The dataset adopts laboratory-measured optical constants, is optimized for the observation requirements of the James Webb Space Telescope (JWST) and the future ARIEL telescope, and supports rapid retrieval of cloud properties. It is primarily applied to exoplanetary atmospheric characterization, addresses the computational bottleneck in aerosol modeling, and facilitates multi-species cloud research and population statistical analysis.




