湖南理工学院学报(自然科学版)2024,Vol.37Issue(3) :25-29.

基于深度学习的甲骨文字识别系统设计

Design of a Deep Learning-based System for Oracle Bone Script Recognition

王峻韬 李辰浩 郑红
湖南理工学院学报(自然科学版)2024,Vol.37Issue(3) :25-29.

基于深度学习的甲骨文字识别系统设计

Design of a Deep Learning-based System for Oracle Bone Script Recognition

王峻韬 1李辰浩 1郑红1
扫码查看

作者信息

  • 1. 华东理工大学 信息科学与工程学院,上海 200237
  • 折叠

摘要

甲骨文作为中国最古老的成熟文字系统,对其识别和学习一直备受各方关注.由于甲骨文字内容多变、拓片噪声多等原因,很少有完整的系统可以对甲骨文字进行识别与学习.将深度学习技术应用于甲骨文字图片识别与学习,可以有效缓解该问题.针对该问题,构建一个基于残差网络模型的系统,将甲骨文字识别、查询和学习相结合.系统不仅能够准确识别甲骨文字,还具备实时查询和学习功能,用户可以通过系统快速获取相关信息并进行知识学习,可为甲骨文字的传承与发展提供技术支持.

Abstract

As the oldest mature writing system in China,the recognition and learning of oracle bone scripts have garnered significant attention from various stakeholders.However,comprehensive systems for the recognition and learning of oracle bone scripts remain limited due to the diverse content of the characters and the noise present in the rubbings.The application of deep learning technology in recognizing and learning from oracle bone script images can effectively mitigate this issue.To address this challenge,we have developed a system based on a residual network model that integrates oracle bone recognition,querying,and learning.This system not only accurately identifies oracle bone scripts but also offers real-time querying and learning functionalities.Users can quickly access relevant information and acquire knowledge through the system,thereby providing technical support for the preservation and advancement of oracle scripts.

关键词

甲骨文字识别/残差网络/文字识别/深度学习

Key words

oracle bone recognition/residual network/character recognition/deep learning

引用本文复制引用

出版年

2024
湖南理工学院学报(自然科学版)
湖南理工学院

湖南理工学院学报(自然科学版)

影响因子:0.259
ISSN:1672-5298
段落导航相关论文