DeepJiandu dataset for character detection and recognition on Jiandu manuscript

In ancient China, bamboo and wooden slips referred to as "Jiandu" served as primary mediums for documenting historical events prior to the advent of paper. These artifacts are rich in historical data and cultural significance. Accurate identification of characters on Jiandu is essential for deciphering the historical narratives they contain and plays a vital role in processing Jiandu manuscripts. We introduce the DeepJiandu dataset, specifically designed for the detection and recognition of Jiandu characters. Comprising 7,416 images annotated with 99,888 characters across 2,272 categories, this dataset addresses a variety of complex challenges encountered in Jiandu character recognition, including character degradation, diverse layouts, and variable forms and shapes. The authenticity and reliability of the DeepJiandu dataset render it an effective tool for training and evaluating models geared towards Jiandu character recognition.

在中国古代，纸出现之前，被称为‘简牍（Jiandu）’的竹片和木片是记录历史事件的主要媒介。这些文物蕴含丰富的历史数据与文化价值。准确识别简牍上的文字对于解读其中包含的历史叙事至关重要，并且在简牍文献处理中发挥着关键作用。我们提出DeepJiandu数据集，该数据集专为简牍文字的检测与识别而设计。该数据集包含7416张图像，标注了2272个类别下的99888个文字，旨在应对简牍文字识别中遇到的多种复杂挑战，包括文字退化、布局多样以及形态变异等。DeepJiandu数据集的真实性与可靠性使其成为训练和评估简牍文字识别模型的有效工具。