Improvement of Oracle Bone Inscription Recognition Accuracy: A Deep Learning Perspective
Deep learning techniques have been successfully applied in handwriting recognition. Oracle bone inscriptions (OBI) are the earliest hieroglyphs in China and valuable resources for studying the etymology of Chinese characters. OBI are of important historical and cultural value in China; thus, textual...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-01-01
|
| Series: | ISPRS International Journal of Geo-Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2220-9964/11/1/45 |
| _version_ | 1797493496300437504 |
|---|---|
| author | Xuanming Fu Zhengfeng Yang Zhenbing Zeng Yidan Zhang Qianting Zhou |
| author_facet | Xuanming Fu Zhengfeng Yang Zhenbing Zeng Yidan Zhang Qianting Zhou |
| author_sort | Xuanming Fu |
| collection | DOAJ |
| description | Deep learning techniques have been successfully applied in handwriting recognition. Oracle bone inscriptions (OBI) are the earliest hieroglyphs in China and valuable resources for studying the etymology of Chinese characters. OBI are of important historical and cultural value in China; thus, textual research surrounding the characters of OBI is a huge challenge for archaeologists. In this work, we built a dataset named OBI-100, which contains 100 classes of oracle bone inscriptions collected from two OBI dictionaries. The dataset includes more than 128,000 character samples related to the natural environment, humans, animals, plants, etc. In addition, we propose improved models based on three typical deep convolutional network structures to recognize the OBI-100 dataset. By modifying the parameters, adjusting the network structures, and adopting optimization strategies, we demonstrate experimentally that these models perform fairly well in OBI recognition. For the 100-category OBI classification task, the optimal model achieves an accuracy of 99.5%, which shows competitive performance compared with other state-of-the-art approaches. We hope that this work can provide a valuable tool for character recognition of OBI. |
| first_indexed | 2024-03-10T01:20:49Z |
| format | Article |
| id | doaj.art-47c9758cfbbe4076be9ec557667bdf96 |
| institution | Directory Open Access Journal |
| issn | 2220-9964 |
| language | English |
| last_indexed | 2024-03-10T01:20:49Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | ISPRS International Journal of Geo-Information |
| spelling | doaj.art-47c9758cfbbe4076be9ec557667bdf962023-11-23T14:00:15ZengMDPI AGISPRS International Journal of Geo-Information2220-99642022-01-011114510.3390/ijgi11010045Improvement of Oracle Bone Inscription Recognition Accuracy: A Deep Learning PerspectiveXuanming Fu0Zhengfeng Yang1Zhenbing Zeng2Yidan Zhang3Qianting Zhou4Software Engineering Institute, East China Normal University, Shanghai 200062, ChinaSoftware Engineering Institute, East China Normal University, Shanghai 200062, ChinaDepartment of Mathematics, Shanghai University, Shanghai 200444, ChinaSoftware Engineering Institute, East China Normal University, Shanghai 200062, ChinaSoftware Engineering Institute, East China Normal University, Shanghai 200062, ChinaDeep learning techniques have been successfully applied in handwriting recognition. Oracle bone inscriptions (OBI) are the earliest hieroglyphs in China and valuable resources for studying the etymology of Chinese characters. OBI are of important historical and cultural value in China; thus, textual research surrounding the characters of OBI is a huge challenge for archaeologists. In this work, we built a dataset named OBI-100, which contains 100 classes of oracle bone inscriptions collected from two OBI dictionaries. The dataset includes more than 128,000 character samples related to the natural environment, humans, animals, plants, etc. In addition, we propose improved models based on three typical deep convolutional network structures to recognize the OBI-100 dataset. By modifying the parameters, adjusting the network structures, and adopting optimization strategies, we demonstrate experimentally that these models perform fairly well in OBI recognition. For the 100-category OBI classification task, the optimal model achieves an accuracy of 99.5%, which shows competitive performance compared with other state-of-the-art approaches. We hope that this work can provide a valuable tool for character recognition of OBI.https://www.mdpi.com/2220-9964/11/1/45cultural heritageoracle bone inscriptionsdeep learningCNNcharacter recognitionimage classification |
| spellingShingle | Xuanming Fu Zhengfeng Yang Zhenbing Zeng Yidan Zhang Qianting Zhou Improvement of Oracle Bone Inscription Recognition Accuracy: A Deep Learning Perspective ISPRS International Journal of Geo-Information cultural heritage oracle bone inscriptions deep learning CNN character recognition image classification |
| title | Improvement of Oracle Bone Inscription Recognition Accuracy: A Deep Learning Perspective |
| title_full | Improvement of Oracle Bone Inscription Recognition Accuracy: A Deep Learning Perspective |
| title_fullStr | Improvement of Oracle Bone Inscription Recognition Accuracy: A Deep Learning Perspective |
| title_full_unstemmed | Improvement of Oracle Bone Inscription Recognition Accuracy: A Deep Learning Perspective |
| title_short | Improvement of Oracle Bone Inscription Recognition Accuracy: A Deep Learning Perspective |
| title_sort | improvement of oracle bone inscription recognition accuracy a deep learning perspective |
| topic | cultural heritage oracle bone inscriptions deep learning CNN character recognition image classification |
| url | https://www.mdpi.com/2220-9964/11/1/45 |
| work_keys_str_mv | AT xuanmingfu improvementoforacleboneinscriptionrecognitionaccuracyadeeplearningperspective AT zhengfengyang improvementoforacleboneinscriptionrecognitionaccuracyadeeplearningperspective AT zhenbingzeng improvementoforacleboneinscriptionrecognitionaccuracyadeeplearningperspective AT yidanzhang improvementoforacleboneinscriptionrecognitionaccuracyadeeplearningperspective AT qiantingzhou improvementoforacleboneinscriptionrecognitionaccuracyadeeplearningperspective |