Chengyuan Song
Papers
1
Total Citations
11
H-Index
1
About
Chengyuan Song’s research focuses on agricultural robotics and computer vision, with a particular emphasis on developing efficient, lightweight models for real-time crop detection and robotic harvesting. His major contribution is the creation of TDPPL-Net, a novel architecture that simultaneously detects tomatoes and localizes picking points with remarkable speed and accuracy. This work directly addresses a critical bottleneck in agricultural automation: the deployment of detection models on low-cost, GPU-free industrial PCs, where traditional large models fail due to computational constraints. By drastically reducing network parameters without sacrificing performance, Song’s approach enables affordable, real-time harvesting robots. His TDPPL-Net paper has already garnered 11 citations, signaling its growing influence in precision agriculture and robotics. Song’s work stands out for its practical impact—bridging the gap between cutting-edge deep learning and real-world agricultural constraints. He is recognized for advancing the field toward more accessible, efficient automation solutions that can empower small-scale farms and reduce labor dependency. His research continues to inspire new directions in lightweight vision models for resource-limited environments.
Research Focus
Key Achievements
Top Papers
- 1