Dynamic Dataset Augmentation for Deep Learning-based Oracle Bone Inscriptions Recognition

Abstract

Oracle bone inscriptions (OBIs) are a kind of hieroglyph, used about 3,600 years ago for divination and the recording of events. The characters on these OBIs are of great interest because they are precursors to the modern Chinese characters widely used across Asia today. However, as the OBIs were only discovered in 1899, there are currently few documents to describe them. Hence, recognizing and unlocking the meaning of OBIs helps to understand the ancient history of China, the evaluation of Chinese characters, and more. Currently, deep learning has made great progress and brought about a revolution in the research field of recognition, and shows good potential to meet the challenges of OBIs recognition. Due to the scarcity of resources, many OBIs contain only a small number of instances, which causes dataset imbalance and limits the accuracy of recognition. This paper attempts to provide a suite of OBIs recognition methods comprising an original OBIs dataset creation, dynamic dataset augmentation, and a novel deep learning-based recognition method. To this end, we create an original OBIs dataset and propose a modified Generative Adversarial Network for augmenting the original OBIs dataset. The augmented data is then dynamically selected for training the deep learning model considering the data imbalance problem. A novel model called C-A Net is proposed for OBIs recognition. The results of evaluation experiments show that the dynamical dataset augmentation can effectively locate a suitable training dataset for the deep learning model and solve the problem of imbalanced OBIs distribution. In addition, the recognition accuracy of C-A Net is 91.10%, which is higher than that of eight state-of-the-art models, and thus effectively suppresses the occurrence of overfitting. We also present an original OBIs dataset named OBI125, which is currently the only rubbing-type OBIs dataset that is open to the public. The code is available at http://www.ihpc.se.ritsumei.ac.jp/obidataset.html.

ViT-based classification of oracle bones for periodization of the Shang Dynasty

“…In recent years, deep-learning-based methods have gained wide attention for Oracle bone script recognition. proposed a dynamic data augmentation method that achieved good recognition performance for imbalanced 125 categories of tablet scripts in their self-built dataset OBI125 [10]. However, the abovementioned studies mainly focused on the significant intra-class differences of Oracle bone script forms, which only included the 100-300 most frequent character categories, and the recognition targets were tablet Oracle bone scripts with relatively large sample sizes.…”

The early Japanese books reorganization by combining image processing and deep learning

“…It also means the OBIs are a prototype of some Japanese characters. For overcoming the problem of the dataset limitation in OBIs recognition, Yue et al., have proposed a Generative Adversarial Network (GAN) based method for data augmentation [23]. This research has achieved a certain level.…”

Oracle recognition based on improved YOLOv7

“…Deep learning mimics the learning hierarchical model structure of the human brain and extracts features from the input data from the lower level to the upper level, so that the corresponding relationship between the lower-level signals and the upper-level information can be effectively established. At present, the state attaches great importance to the research of oracle bones [5][6][7], but the current research of oracle bones by researchers at home and abroad does not meet the expected requirements in terms of oracle identification accuracy, time and the ability of the model to prevent interference. Except for the researches of relevant experts, the information about oracle bones is hardly known by the public.…”