Dongxu Gao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Dongxu Gao has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Yalin Zheng
Department of Eye and Vision Science, Institute of Life Course and Medical Sciences University of Liverpool Liverpool UK
2 shared publications Amira Stylianides
St. Pauls Eye Unit, Royal Liverpool University Hospital
1 shared publication- Boyu Hu
Guangxi Key Laboratory of Intelligent Control and Maintenance of Power Equipment, School of Electrical Engineering, Guangxi University, Nanning, China
1 shared publication Bryan M. Williams
University of Liverpool
1 shared publication- Dalin Zhou
University of Portsmouth
1 shared publication - Davide Borroni
Riga Stradins University, Riga, Latvia
1 shared publication
Latest Publications
- 10 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-14
Simplify-YOLOv5m: A Simplified High-Speed Insulator Detection Algorithm for UAV Images
As an essential power device, insulators are present in large numbers across various transmission lines. Insulators are continuously affected by the natural environment, leading to fractures, pollution accumulation, and other issues. These problems can cause flashovers and pollution flash discharge accidents, trigger relay protection devices, and result in significant economic losses. However, it is challenging for existing algorithms to achieve a good balance between accuracy and speed in many insulator datasets. Therefore, in this paper, we construct a novel feature extraction module named the point-depth module (PD module) to reduce feature loss and computational complexity. We also design the ShortPath-FPN (SP-FPN), which introduces the Attentional Feature Fusion module (AFF) to effectively shorten the information flow from input to head and enhance feature maps. Compared to state-of-the-art (SOTA) algorithms, the PD module significantly reduces the amount of computation with only a slight impact on accuracy, while the SP-FPN module effectively improves both accuracy and detection speed. We enhance the YOLOV5m model based on the above modules to achieve an accuracy of 94.1% and an FPS of 77 for insulator detection, meeting the requirements for real-time detection while improving detection accuracy and outperforming other methods.
- 10 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-14
Simplify-YOLOv5m: A Simplified High-Speed Insulator Detection Algorithm for UAV Images
As an essential power device, insulators are present in large numbers across various transmission lines. Insulators are continuously affected by the natural environment, leading to fractures, pollution accumulation, and other issues. These problems can cause flashovers and pollution flash discharge accidents, trigger relay protection devices, and result in significant economic losses. However, it is challenging for existing algorithms to achieve a good balance between accuracy and speed in many insulator datasets. Therefore, in this paper, we construct a novel feature extraction module named the point-depth module (PD module) to reduce feature loss and computational complexity. We also design the ShortPath-FPN (SP-FPN), which introduces the Attentional Feature Fusion module (AFF) to effectively shorten the information flow from input to head and enhance feature maps. Compared to state-of-the-art (SOTA) algorithms, the PD module significantly reduces the amount of computation with only a slight impact on accuracy, while the SP-FPN module effectively improves both accuracy and detection speed. We enhance the YOLOV5m model based on the above modules to achieve an accuracy of 94.1% and an FPS of 77 for insulator detection, meeting the requirements for real-time detection while improving detection accuracy and outperforming other methods.
- 25 May 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 24 (12) , 3480-3490
Cooperative Low-Rank Models for Removing Stripe Noise From OCTA Images
Optical coherence tomography angiography (OCTA) is an emerging non-invasive imaging technique for imaging the microvasculature of the eye based on phase variance or amplitude decorrelation derived from repeated OCT images of the same tissue area. Stripe noise occurs during the OCTA acquisition process due to the involuntary movement of the eye. To remove the stripe noise (or 'destriping') effectively, we propose two novel image decomposition models to simultaneously destripe all the OCTA images of the same eye in a cooperative way: cooperative uniformity destriping (CUD) and cooperative similarity destriping (CSD) model. Both the models consider stripe noise by low-rank constraint but in different ways: the CUD model assumes that stripe noise is identical across all the images while the CSD model assumes that the stripe noise at different layers are different and have to be considered in the model. Compared to the CUD model, CSD is a more general solution for real OCTA images. An efficient solution (CSD+) is developed for model CSD to reduce the computational complexity. The models were extensively evaluated against state of the art methods on both synthesized and real OCTA images. The experiments demonstrated not only the effectiveness of the CSD and CSD+ models in terms of peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) and CSD+ is twice faster than CSD, but also their beneficiary effect on the vessel segmentation of OCTA images. We expect our models will become a powerful tool for clinical applications.
- 25 May 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 24 (12) , 3480-3490
Cooperative Low-Rank Models for Removing Stripe Noise From OCTA Images
Optical coherence tomography angiography (OCTA) is an emerging non-invasive imaging technique for imaging the microvasculature of the eye based on phase variance or amplitude decorrelation derived from repeated OCT images of the same tissue area. Stripe noise occurs during the OCTA acquisition process due to the involuntary movement of the eye. To remove the stripe noise (or 'destriping') effectively, we propose two novel image decomposition models to simultaneously destripe all the OCTA images of the same eye in a cooperative way: cooperative uniformity destriping (CUD) and cooperative similarity destriping (CSD) model. Both the models consider stripe noise by low-rank constraint but in different ways: the CUD model assumes that stripe noise is identical across all the images while the CSD model assumes that the stripe noise at different layers are different and have to be considered in the model. Compared to the CUD model, CSD is a more general solution for real OCTA images. An efficient solution (CSD+) is developed for model CSD to reduce the computational complexity. The models were extensively evaluated against state of the art methods on both synthesized and real OCTA images. The experiments demonstrated not only the effectiveness of the CSD and CSD+ models in terms of peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) and CSD+ is twice faster than CSD, but also their beneficiary effect on the vessel segmentation of OCTA images. We expect our models will become a powerful tool for clinical applications.
- 24 January 2020
- Published by Springer Nature in Communications in Computer and Information Science
- p. 189-197
A Novel Deep Learning Based OCTA De-striping Method
- 24 January 2020
- Published by Springer Nature in Communications in Computer and Information Science
- p. 189-197
A Novel Deep Learning Based OCTA De-striping Method