Rosa Pia Devanna

About

Rosa Pia Devanna is a leading researcher at the intersection of precision agriculture, robotics, and computer vision. Her work focuses on developing automated, vision-based systems for in-field crop monitoring, fruit detection, and yield estimation—critical tasks for modern, data-driven farming. Devanna’s major contributions include pioneering semi-supervised deep learning frameworks for fruit identification, such as her work on pomegranate detection using a farmer robot (24 citations), and advancing grapevine phenotyping by combining deep segmentation with depth-based clustering for yield estimation in precision viticulture (11 citations). She has also tackled challenging detection scenarios for tomatoes using YOLO-based single-stage detectors (5 citations) and developed novel pipelines for automated grape bunch detection and counting using RGB-D data from a farmer robot (4 citations). Notably, her research explores the use of consumer-grade RGB-D cameras, like the Intel RealSense D435, to extract crop health indicators such as NDVI, making precision agriculture more accessible and scalable. With a growing citation record and a focus on real-world deployment, Devanna is shaping the future of autonomous agricultural systems.

Research Focus

Key Achievements

Top Papers

1. 1
   In-Field Automatic Identification of Pomegranates Using a Farmer Robot

   24 citations · 2022
2. 2
   Yield estimation in precision viticulture by combining deep segmentation and depth-based clustering

   11 citations · 2025
3. 3
   Tomato detection in challenging scenarios using YOLO-based single stage detectors

   5 citations · 2023
4. 4
   Automated detection and counting of grape bunches using a farmer robot

   4 citations · 2023
5. 5
   Automated plant-scale crop monitoring by a farmer robot using a consumer-grade RGB-D camera

   2 citations · 2023

Key Collaborators

Contact & Links