Deyong Liu

Weifang University of Science and Technology, Concordia University

Papers

Total Citations

H-Index

About

Deyong Liu is a researcher at the forefront of agricultural robotics and intelligent automation, with a primary focus on computer vision and deep learning for precision agriculture. His work addresses critical challenges in automated fruit picking, particularly the accurate detection of crops under complex environmental conditions such as variable lighting, occlusion, and overlapping fruits. In his highly cited 2024 study, Liu introduced an improved YOLOv7 model enhanced by a Swin Transformer and Trident Pyramid Networks, achieving state-of-the-art performance in tomato detection—a breakthrough that directly impacts the commercial viability of robotic harvesting systems. Earlier, Liu pioneered the application of deep learning to manipulator visual positioning, developing a method that enables robots to recognize and localize unknown objects for precise visual servoing. This foundational work, published in 2018, demonstrated how convolutional neural networks could bridge the gap between perception and robotic control. With over 11 citations across his key publications, Liu’s research is gaining traction among engineers and agritech developers seeking robust, real-time solutions for autonomous fruit picking and industrial manipulation.