Shuyu Fan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Shuyu Fan has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Bohan Li
School of Computer Science, Qufu Normal University, Rizhao 276800, China
2 shared publications - Tao Li
School of Computer Science, Qufu Normal University, Rizhao 276800, China
2 shared publications - Gang Cao
School of Mangement Science and Engineering, Shandong University of Finance and Economics, Jinan, China
1 shared publication - Gang Cao
School of Mangement Science and Engineering, Shandong University of Finance and Economics, Jinan 250014, China
1 shared publication Kholoud Khaled
College of Artificial Intelligence, China University of Petroleum
1 shared publication- Kui Lu
School of Computer Science, Qufu Normal University, Rizhao, China
1 shared publication
Latest Publications
- Open Access
- 4 December 2025
- Published by MDPI AG in Information
- Vol. 16 (12) , 1072
OTSU-UCAN: An OTSU-Based Integrated Satellite–Terrestrial Information System for 6G in Vehicle Navigation
One of the bottlenecks hindering the applications (e.g., vehicle navigation) of blockchain–UCAN is privacy. A sharded blockchain can protect vehicle data to a certain extent. However, unbalanced shard loads lead to low throughput and poor feature extraction in blockchain–UCAN. This paper proposes an optimal image binarization method (OTSU-GK) to enhance image features and reduce the amount of uploaded data, thereby improving throughput. Specifically, OTSU-GK uses a Gaussian kernel method where the parameters are optimized using grid search to improve the calculation of the threshold. Additionally, we have designed a Node Load Score (NLS)-based sharding blockchain, which considers the number of historical transactions, transaction types, transaction frequency, and other metrics to balance the sharding loads and further improve throughput. The experimental results show that OTSU-GK improves by 74.3%, 58.7%, and 83% in SSIM, RMSE/MAE/AER, and throughput. In addition, it reduces IL by 70.3% compared to other methods.
- 4 December 2025
- Published by MDPI AG in Information
- Vol. 16 (12) , 1072
Open AccessOTSU-UCAN: An OTSU-Based Integrated Satellite–Terrestrial Information System for 6G in Vehicle Navigation
One of the bottlenecks hindering the applications (e.g., vehicle navigation) of blockchain–UCAN is privacy. A sharded blockchain can protect vehicle data to a certain extent. However, unbalanced shard loads lead to low throughput and poor feature extraction in blockchain–UCAN. This paper proposes an optimal image binarization method (OTSU-GK) to enhance image features and reduce the amount of uploaded data, thereby improving throughput. Specifically, OTSU-GK uses a Gaussian kernel method where the parameters are optimized using grid search to improve the calculation of the threshold. Additionally, we have designed a Node Load Score (NLS)-based sharding blockchain, which considers the number of historical transactions, transaction types, transaction frequency, and other metrics to balance the sharding loads and further improve throughput. The experimental results show that OTSU-GK improves by 74.3%, 58.7%, and 83% in SSIM, RMSE/MAE/AER, and throughput. In addition, it reduces IL by 70.3% compared to other methods.
- Open Access
- 4 November 2025
- Published by MDPI AG in Preprints.org
OTSU-UCAN: An OTSU-Based Integrated Satellite-Terrestrial Information System for 6G in Vehicle Navigation
Privacy is one of the bottlenecks that hinder the applications (e.g. vehicle navigation) of blockchain-UCAN. A sharded blockchain protects the privacy of vehicle data to a certain extent. However, the unbalanced shard loads leads to low throughput, and the feature extraction in blockchain-UCAN is still poor. In this paper, an optimal image binarisation method OTSU-GK is proposed to enhance image features while reducing the amount of uploaded data to further improve the throughput. Specifically, OTSU-GK employs a Gaussian kernel method, where the parameters are optimized through grid search, to optimize the calculation of the threshold. In addition, we design a Node Load Score-based (NLS) sharding blockchain, which considers the number of historical transactions, transaction types, transaction frequency and other metrics, to balance the sharding loads and further improve the throughput. The experimental results show that OTSU-GK pproximately a 50% improvement in metrics and 83% improvement in throughput compared to other methods.
- 4 November 2025
- Published by MDPI AG in Preprints.org
Open AccessOTSU-UCAN: An OTSU-Based Integrated Satellite-Terrestrial Information System for 6G in Vehicle Navigation
Privacy is one of the bottlenecks that hinder the applications (e.g. vehicle navigation) of blockchain-UCAN. A sharded blockchain protects the privacy of vehicle data to a certain extent. However, the unbalanced shard loads leads to low throughput, and the feature extraction in blockchain-UCAN is still poor. In this paper, an optimal image binarisation method OTSU-GK is proposed to enhance image features while reducing the amount of uploaded data to further improve the throughput. Specifically, OTSU-GK employs a Gaussian kernel method, where the parameters are optimized through grid search, to optimize the calculation of the threshold. In addition, we design a Node Load Score-based (NLS) sharding blockchain, which considers the number of historical transactions, transaction types, transaction frequency and other metrics, to balance the sharding loads and further improve the throughput. The experimental results show that OTSU-GK pproximately a 50% improvement in metrics and 83% improvement in throughput compared to other methods.
- 3 November 2024
- Published by Springer Nature
- p. 93-113
ML-SCODNet: Multitask Learning for Scene Classification and Object Detection Network from Remote Sensing Images
- 3 November 2024
- Published by Springer Nature
- p. 93-113
ML-SCODNet: Multitask Learning for Scene Classification and Object Detection Network from Remote Sensing Images