Abstract:Image segmentation and recognition of oracle bone topography is a common challenge for in-depth study of oracle bones. The key lies in how to accurately and efficiently process and recognize the original oracle bone topography images and solve the practical problems in the research based on the image digitization results. The research of this topic helps to promote the research of oracle bone holography and the construction of digitization project. Image segmentation of oracle bone tablet is to use digital image processing and computer vision technology to extract independent text regions with distinctive image features and non-overlapping each other in the complex background of the original topographic image. The study of automatic segmentation and recognition of oracle bone characters on original topos has important social significance. In this paper, the image preprocessing model based on the channel attention mechanism is firstly utilized to initially judge and identify the interference elements present in the original oracle bone topography image. Meanwhile, a joint filtering noise reduction model is established to effectively remove the location-like noise and artificial texture in the image. Then, a fast oracle bone image segmentation model based on neural network is established, and each single character region in the image is successfully segmented out by building and training the model through deep learning. In this paper, 76 common oracle scripts and their variant scripts are trained and recognized in combination with the oracle recognition enhancement model, and a one-to-many link between Chinese characters and oracle images is established. The accuracy and efficiency of oracle bone recognition are further improved by segmenting and classifying the test set images for recognition.

Method Combining Rule-Based And Corpus-Based Approaches For Oracle-Bone Inscription Information Processing

Incorporating External POS Tagger for Punctuation Restoration

A study on encoding-based oracle bone script recognition

A graphic-based method for Chinese Oracle-bone classification

Research on automatic segmentation and recognition of original topographic single characters based on intelligent recognition of oracle bones

A Method Combining Rule-based and Statistics-based Approaches for Chinese Word Segmentation

Image Segmentation and Recognition of Oracle Bone Topographies Based on Deep Learning

Blending segmentation with tagging in Chinese language corpus processing

OracleSage: Towards Unified Visual-Linguistic Understanding of Oracle Bone Scripts through Cross-Modal Knowledge Fusion

Study on Japanese Word Segmentation and POS Tagging Based on Rules and Statistics

Oracle Bone Script Intelligent Recognition: Automatic Segmentation and Recognition of Original Rubbing Single Characters

OBI-Bench: Can LMMs Aid in Study of Ancient Script on Oracle Bones?

Deciphering Oracle Bone Language with Diffusion Models

Oracle Bone Script Similiar Character Screening Approach Based on Simsiam Contrastive Learning and Supervised Learning

Automatic Segmentation of Oracle Bone Inscriptions Using YOLOv8

Joint Chinese Word Segmentation and POS Tagging on Heterogeneous Annotated Corpora with Multiple Task Learning.

Automated Recognition of Oracle Bone Inscriptions Using Deep Learning and Data Augmentation

Classical Chinese Sentence Segmentation for Tomb Biographies of Tang Dynasty

Research on Corpus Annotation Method Based on Collective Intelligence

Applying Deep Learning Algorithms for Automatic Recognition and Transcription of Texts in Oracle Bones and Golden Texts

A Sentence Segmentation Method for Ancient Chinese Texts Based on NNLM.