October 2025 |
Dr. Hongkai Yu, an associate professor in the Department of Electrical and Computer Engineering (ECE), will join researchers from the University of Kentucky to apply computer vision and artificial intelligence to the task of improving crop quality and yields in large scale greenhouses. The work is funded by a nearly $1.2 million grant from the National Science Foundation, titled “An Autonomous Robotic System for Precision and High-Throughput Tomato Phenotyping in Large-Scale Greenhouses.”
Dr. Yu and his students will design computer vision and AI algorithms that use camera imaging of the plants for phenotyping. Phenotypes (visible/observable traits) like fruit size/weight, number of fruit, and plant architecture can be used to optimize yield. The computer vision/AI system will improve the speed and quality of plant data collection and analysis while also reducing potential work hazards for greenhouse workers.
Dr. Yu directs the Cleveland Vision & AI Lab, which pursues innovative research in the areas of computer vision, machine learning, deep learning and artificial intelligence. His lab recently published CSU’s first paper in the high impact journal Computer Vision and Pattern Recognition (CVPR). CSU PhD students Jinlong Li and Xinyu Liu (pictured below) presented the paper at the 2024 CVPR conference.
Jinlong Li, Baolu Li, Zhengzhong Tu, Xinyu Liu, Qing Guo, Felix Juefei-Xu, Runsheng Xu, Hongkai Yu. Light the Night: A Multi-Condition Diffusion Framework for Unpaired Low-Light Enhancement in Autonomous Driving. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024.
UPDATE: Invention Disclosure Submission An invention disclosure was submitted by Dr. Zhiqiang Gao, an associate professor in the Department of Electrical and Computer Engineering and Director of the Center for Advanced Control Technologies (CACT), and his PhD student Yu Hu, titled “A Model-Free Pid Tuning Method.” The method makes controller tuning completely intelligible to average users, thus saving energy and improving performance. It provides a comprehensive solution for all industrial control design and tuning problems using the same set of PID gains, but pins them down in terms of the bandwidth, a concept most control engineers with some training are familiar. |
CSU Licenses IP to ComHealth The CSU Technology Transfer Office has negotiated and secured a license agreement with ComHealth Data Solutions, LLC for US Patent Application No. 17/072,984 titled “System and Method for Improving Healthcare Through Community Engagement” and the associated copyrights. The I-Hope platform is a first-in-class, SaaS based system for easing and streamlining the client health data collection process for frontline public health workers. ComHealth was founded by Dr. William Matcham, an associate professor in the School of Nursing, and recently received a $200,000 grant from the Ohio Third Frontier to continue its commercialization efforts. |
© 2025 - Cleveland State University - Office of Research |
