深層生成モデルを用いた中国出土木簡の書体分析

Material type: 図書

Author: 吉田, 壮

Publisher: 関西大学アジア・オープン・リサーチセンター

Publication date: 2022-03-31

Material Format: Digital

Capacity, size, etc.: -

NDC: -

View All

Notes on use

Note (General)：: In this paper, we develop a method to analyze the calligraphic styles of wood slips excavated in Northwestern China. Specifically, we propose a method...

Search by Bookstore

Holdings of Libraries in Japan

This page shows libraries in Japan other than the National Diet Library that hold the material.

List of Cooperating Institutions and Databases

Please contact your local library for information on how to use materials or whether it is possible to request materials from the holding libraries.

other

Kansai University Repository
Digital
You can check the holdings of institutions and databases with which Institutional Repositories DataBase(IRDB)(Institutional Repository) is linked at the site of Institutional Repositories DataBase(IRDB)(Institutional Repository).
Kansai University Repository

Search by Bookstore

Bibliographic Record

You can check the details of this material, its authority (keywords that refer to materials on the same subject, author's name, etc.), etc.

Digital

Material Type: 図書
Title: 深層生成モデルを用いた中国出土木簡の書体分析
Author/Editor: 吉田, 壮
Author Heading: 吉田, 壮
Publication, Distribution, etc.: 関西大学アジア・オープン・リサーチセンター
Publication Date: 2022-03-31
Publication Date (W3CDTF): 2022-03-31
Alternative Title: Analysis of Writing Styles on Wood Slips of the Chinese Han Period Using Deep Generative Models
Periodical title: KU-ORCASが開くデジタル化時代の東アジア文化研究 : オープン・プラットフォームで浮かび上がる、新たな東アジアの姿
Pages: 295-305
Text Language Code: jpn
Subject Heading: 木簡
書体分析
Generative Adversarial Network
関西大学
Kansai University
Target Audience: 一般
Note (General): In this paper, we develop a method to analyze the calligraphic styles of wood slips excavated in Northwestern China. Specifically, we propose a method to quantify the degree of collapse of the Chinese characters by measuring the degree of dissociation using a deep generative model. To realize the method, we introduce anomaly detection with generative adversarial networks (AnoGAN), trained by normal data, and then detect abnormal data based on the reconstruction error when those data are input. First, we train the GAN by using images of a character written in Clerical-Script as training data. Next, we calculate the anomaly scores of the characters based on the difference between the Cursive-Script characters and the generated character images and then calculate the degree of collapse. In our experiments, we created datasets consisting of wood slips made in the Han Period and showed that the proposed method could quantify the degree of misalignment between Clerical-Script of neat font and Cursive-Script of scrawl font.
DOI: 10.32286/00026600
Access Restrictions: 限定公開
Data Provider (Database): 国立情報学研究所 : 学術機関リポジトリデータベース（IRDB）（機関リポジトリ）
https://irdb.nii.ac.jp
Original Data Provider (Database): 関西大学 : 関西大学学術リポジトリ