Shigemichi Matsuzaki
Papers
6
Total Citations
65
H-Index
4
About
Shigemichi Matsuzaki is a pioneering researcher in agricultural robotics, specializing in visual navigation and semantic mapping for unstructured, plant-rich environments. His work addresses a critical gap in autonomous mobile robots: navigating spaces where traversable paths are obscured by foliage, branches, and other vegetation. Matsuzaki’s major contributions include developing robust 3D semantic mapping methods for greenhouses, enabling robots to distinguish between obstacles and traversable plants using trajectory-based object recognition. He also introduced innovative image-based scene recognition techniques that estimate plant traversability without manual annotation, significantly reducing the labor-intensive training process. His multi-source pseudo-label learning approach for semantic segmentation leverages diverse public datasets to train models for unfamiliar greenhouse environments, achieving high accuracy with minimal labeled data. With over 65 citations across his key papers, Matsuzaki’s impact is evident in advancing practical, cost-effective solutions for agricultural automation. Notable achievements include his end-to-end path estimation framework and automatic dataset generation methods, which streamline deep visual navigation in unstructured settings. His work is essential reading for researchers and students interested in field robotics, computer vision, and sustainable agriculture.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6