Defu Yang

Shanghai Civil Aviation College, UCSI University

Papers

Total Citations

H-Index

About

Defu Yang is a researcher advancing the intersection of computer vision and robotics for intelligent retail and smart city applications. His work centers on developing efficient, real-time perception systems, with key contributions in lightweight deep learning models and multi-sensor fusion for autonomous navigation. Yang’s most cited paper, “LSR-YOLO: A lightweight and fast model for retail products detection” (2025, 6 citations), introduces a streamlined object detection framework that significantly reduces computational overhead while maintaining high accuracy, enabling practical deployment in resource-constrained retail environments. Complementing this, his work “Improved Localization Algorithm Based on Multi-Sensor Fusion for Shopping Robots” (2024, 2 citations) tackles critical challenges in indoor robotics, such as arbitrary-start localization and map mismatch, by fusing LiDAR and RGB-D data within an enhanced ORB-SLAM3 system. This innovation directly supports the reliable navigation of shopping robots in dynamic, cluttered spaces. Yang’s research is notable for its focus on bridging algorithmic efficiency with real-world deployment, offering scalable solutions that drive the integration of intelligent systems into everyday urban life. His contributions are shaping the future of automated retail and service robotics.