Hong Cheng Source Confirmed
Affiliation confirmed via AI analysis of OpenAlex, ORCID, and web sources.
Researcher
Southern Arkansas University
faculty
Research Areas
Links
Is this your profile? Verify and claim your profile
Biography and Research Information
OverviewAI-generated summary
Hong Cheng's research explores advanced computational methods and their application across diverse scientific fields. Recent work includes the development of mutual graph learning and uncertainty-guided transformer reasoning for camouflaged object detection. Cheng has also investigated deep learning techniques, specifically dilated convolution and attention, for millimeter wave path loss modeling in 5G communications.
Further research interests span medical applications, with publications on exoskeleton-assisted walking for individuals with spinal cord injury, in vitro fluidic systems for studying endothelial cell shear stress, and a nomogram model for predicting postpartum stress urinary incontinence. Cheng has also applied machine learning to predict phase formation in high entropy alloys.
Metrics
- h-index: 5
- Publications: 28
- Citations: 172
Selected Publications
- Skin Lesion Segmentation Using Unet With A Topology Term in Loss Function (2025) DOI
- Multi-Class Label Detection and Bounding Box Regression Using Transformer with a Customized Loss Function (2024) DOI
- Object Localization Using Vision Transformer with a Loss Function Based on IOU and Mean Squared Error (2023) DOI
- A Topological Data Analysis-Based Approach to Object Localization: A Comparison with ViT and Yolov7 (2023) DOI
Collaborators
Researchers in the database who share publications
Similar Researchers
Based on overlapping research topics