LION

LION数据集是首个包含Astrid Lindgren原创草稿和手写内容的数据集，专为手写速记识别研究设计。该数据集包含Lindgren的知名儿童小说《狮子兄弟》等作品的摘录，共计198页，分为2900个单独的图像，适用于手写文本识别和文档图像处理。LION数据集的创建旨在促进手写速记识别的研究，通过结合领域知识和预训练技术，提高识别准确性。该数据集的应用领域包括文学研究、儿童文学分析以及手写文本的自动识别，旨在解决手写速记识别的挑战，推动文化遗产的数字化和保护。

The LION Dataset is the first dataset containing original drafts and handwritten content by Astrid Lindgren, designed specifically for handwritten shorthand recognition research. It includes excerpts from Lindgren's well-known children's novel *The Lion Brothers* and other works, totaling 198 pages and split into 2900 individual images, applicable to handwritten text recognition and document image processing. The LION Dataset was developed to advance research on handwritten shorthand recognition, aiming to improve recognition accuracy by integrating domain knowledge and pre-trained technologies. Its application fields cover literary research, children's literature analysis, and automatic handwritten text recognition, with the goals of addressing the challenges in handwritten shorthand recognition and promoting the digitization and preservation of cultural heritage.