DKDS: A Benchmark Dataset of Degraded Kuzushiji Documents with Seals for Detection and Binarization

TL;DR

The paper introduces DKDS, a new dataset for detecting degraded Kuzushiji characters and seals, addressing noise challenges in OCR tasks with baseline results and available code.

Contribution

It provides the first dedicated dataset for Kuzushiji document degradation, seals, and binarization, along with benchmark tasks and baseline results.

Findings

Baseline YOLO models achieve detection benchmarks.

Traditional and GAN-based methods show varying effectiveness in binarization.

The dataset and code are publicly available for further research.

Abstract

Kuzushiji, a pre-modern Japanese cursive script, can currently be read and understood by only a few thousand trained experts in Japan. With the rapid development of deep learning, researchers have begun applying Optical Character Recognition (OCR) techniques to transcribe Kuzushiji into modern Japanese. Although existing OCR methods perform well on clean pre-modern Japanese documents written in Kuzushiji, they often fail to consider various types of noise, such as document degradation and seals, which significantly affect recognition accuracy. To the best of our knowledge, no existing dataset specifically addresses these challenges. To address this gap, we introduce the Degraded Kuzushiji Documents with Seals (DKDS) dataset as a new benchmark for related tasks. We describe the dataset construction process, which involves the assistance of a trained Kuzushiji expert, and define two benchmark tracks: (1) Kuzushiji character and seal detection and (2) document binarization. For the Kuzushiji character and seal detection track, we provide baseline results using several recent versions of YOLO to detect Kuzushiji characters and seals. For the document binarization track, we present baseline results from traditional binarization algorithms, traditional algorithms combined with K-means clustering, two state-of-the-art (SOTA) generative adversarial network (GAN) methods, and our improved conditional GAN (cGAN)-based method. The DKDS dataset and the implementation code for baseline methods are available at https://ruiyangju.github.io/DKDS.