Enhanced oracle bone corrosion detection using attention-guided YOLO with ghost convolution
Abstract Oracle bone inscriptions (OBIs), as the important records of early Chinese characters, possess profound cultural and historical significance. However, These fragments are susceptible to further damage due to corrosion. Consequently, the identification and localization of corroded regions on...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2024-12-01
|
| Series: | Discover Artificial Intelligence |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s44163-024-00178-5 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832585582565392384 |
|---|---|
| author | Feng Gao Ziheng Yang Qiyu Liu Zhan Zhang Bang Li Han Zhang |
| author_facet | Feng Gao Ziheng Yang Qiyu Liu Zhan Zhang Bang Li Han Zhang |
| author_sort | Feng Gao |
| collection | DOAJ |
| description | Abstract Oracle bone inscriptions (OBIs), as the important records of early Chinese characters, possess profound cultural and historical significance. However, These fragments are susceptible to further damage due to corrosion. Consequently, the identification and localization of corroded regions on these fragments to prevent further deterioration has emerged as a critical task. Current object detection algorithms exhibit limitations in identifying corroded regions on oracle bone fragments, which is manifested by suboptimal detection accuracy, F1 scores, and average precision (AP) values. This deficiency hinders their capability to effectively manage the complexity and diversity of corrosion patterns present on oracle bone fragments. To tackle this challenge, this study incorporates advanced attention mechanisms, including Squeeze-and-Excitation Networks (SE), Coordinate attention, and Convolutional Block Attention Module (CBAM), into the YOLOv5 detection architecture. Additionally, we introduce Ghost convolution into the backbone network to effectively retain critical feature map information while optimizing computational efficiency. Our experimental results indicate that the integration of CBAM attention and Ghost convolution within the YOLOv5 backbone markedly improves the detection accuracy of corroded regions on oracle bone fragments. Specifically, compared to the baseline YOLOv5 model, the proposed models incorporating SE, CBAM, and Ghost convolution obtain the improvements of 2.2%, 6.3%, and 8.5% in AP, respectively. This improvement in detection performance not only facilitates the identification of corroded areas but also contributes to the preservation of oracle bone heritage and the continuity of cultural knowledge. |
| format | Article |
| id | doaj-art-a57f52220e6d4d7aa98edc582ffab1a9 |
| institution | Kabale University |
| issn | 2731-0809 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Springer |
| record_format | Article |
| series | Discover Artificial Intelligence |
| spelling | doaj-art-a57f52220e6d4d7aa98edc582ffab1a92025-01-26T12:42:59ZengSpringerDiscover Artificial Intelligence2731-08092024-12-014111310.1007/s44163-024-00178-5Enhanced oracle bone corrosion detection using attention-guided YOLO with ghost convolutionFeng Gao0Ziheng Yang1Qiyu Liu2Zhan Zhang3Bang Li4Han Zhang5Key Laboratory of Oracle Bone Inscriptions Information Processing, Ministry of EducationSchool of Computer & Information Engineering, Anyang Normal UniversityKey Laboratory of Oracle Bone Inscriptions Information Processing, Ministry of EducationKey Laboratory of Oracle Bone Inscriptions Information Processing, Ministry of EducationKey Laboratory of Oracle Bone Inscriptions Information Processing, Ministry of EducationKey Laboratory of Oracle Bone Inscriptions Information Processing, Ministry of EducationAbstract Oracle bone inscriptions (OBIs), as the important records of early Chinese characters, possess profound cultural and historical significance. However, These fragments are susceptible to further damage due to corrosion. Consequently, the identification and localization of corroded regions on these fragments to prevent further deterioration has emerged as a critical task. Current object detection algorithms exhibit limitations in identifying corroded regions on oracle bone fragments, which is manifested by suboptimal detection accuracy, F1 scores, and average precision (AP) values. This deficiency hinders their capability to effectively manage the complexity and diversity of corrosion patterns present on oracle bone fragments. To tackle this challenge, this study incorporates advanced attention mechanisms, including Squeeze-and-Excitation Networks (SE), Coordinate attention, and Convolutional Block Attention Module (CBAM), into the YOLOv5 detection architecture. Additionally, we introduce Ghost convolution into the backbone network to effectively retain critical feature map information while optimizing computational efficiency. Our experimental results indicate that the integration of CBAM attention and Ghost convolution within the YOLOv5 backbone markedly improves the detection accuracy of corroded regions on oracle bone fragments. Specifically, compared to the baseline YOLOv5 model, the proposed models incorporating SE, CBAM, and Ghost convolution obtain the improvements of 2.2%, 6.3%, and 8.5% in AP, respectively. This improvement in detection performance not only facilitates the identification of corroded areas but also contributes to the preservation of oracle bone heritage and the continuity of cultural knowledge.https://doi.org/10.1007/s44163-024-00178-5Oracle boneCorrosion detectionYOLO algorithmAttention mechanismGhost convolution |
| spellingShingle | Feng Gao Ziheng Yang Qiyu Liu Zhan Zhang Bang Li Han Zhang Enhanced oracle bone corrosion detection using attention-guided YOLO with ghost convolution Discover Artificial Intelligence Oracle bone Corrosion detection YOLO algorithm Attention mechanism Ghost convolution |
| title | Enhanced oracle bone corrosion detection using attention-guided YOLO with ghost convolution |
| title_full | Enhanced oracle bone corrosion detection using attention-guided YOLO with ghost convolution |
| title_fullStr | Enhanced oracle bone corrosion detection using attention-guided YOLO with ghost convolution |
| title_full_unstemmed | Enhanced oracle bone corrosion detection using attention-guided YOLO with ghost convolution |
| title_short | Enhanced oracle bone corrosion detection using attention-guided YOLO with ghost convolution |
| title_sort | enhanced oracle bone corrosion detection using attention guided yolo with ghost convolution |
| topic | Oracle bone Corrosion detection YOLO algorithm Attention mechanism Ghost convolution |
| url | https://doi.org/10.1007/s44163-024-00178-5 |
| work_keys_str_mv | AT fenggao enhancedoraclebonecorrosiondetectionusingattentionguidedyolowithghostconvolution AT zihengyang enhancedoraclebonecorrosiondetectionusingattentionguidedyolowithghostconvolution AT qiyuliu enhancedoraclebonecorrosiondetectionusingattentionguidedyolowithghostconvolution AT zhanzhang enhancedoraclebonecorrosiondetectionusingattentionguidedyolowithghostconvolution AT bangli enhancedoraclebonecorrosiondetectionusingattentionguidedyolowithghostconvolution AT hanzhang enhancedoraclebonecorrosiondetectionusingattentionguidedyolowithghostconvolution |