Shubham Rana
Papers
1
Total Citations
11
H-Index
1
About
Shubham Rana is pioneering the intersection of computer vision and precision agriculture, with a primary focus on robust object detection in complex viticultural environments. His most impactful work, "From vineyard to vision," introduces a multi-domain framework that systematically addresses the persistent challenge of grape cluster detection failures. Rana’s key contribution lies in his cross-domain validation strategy, where he conducted the first comprehensive evaluation of six modern detectors across stratified RGB and NIR datasets spanning vineyards in China, Brazil, and Italy. This work demonstrates that detection performance degrades significantly under real-world conditions such as poor illumination, occlusion, and cluster rotation. To mitigate these failures, Rana developed Oriented YOLOv8-OBB, an orientation-aware detector that reduces false cluster rates by 25–30% in challenging scenarios compared to conventional horizon-based methods. With 11 citations since its 2025 publication, this research is already influencing the design of more resilient agricultural AI systems. Rana’s work is notable for its rigorous cross-dataset methodology and practical impact, offering a blueprint for deploying vision systems that can adapt to the messy, variable conditions of actual vineyards rather than idealized lab settings.
Research Focus
Key Achievements
Top Papers
- 1