This is a set of Python codes for the analysis of Wasan Document using computer vision and deep learning. The main focus is Kanji Detection but there is also interest in Noise reduction and Kanji Classification.
Please stay tunned for news. The plan is to submit the current work to a journal around autumn 2025 and to make the data used publicly available at that time. Open an issue in this repository if you would like early access to the data.