Related Experiment Video
Updated: Jun 3, 2025
05:12
Robotized Testing of Camera Positions to Determine Ideal Configuration for Stereo 3D Visualization of Open-Heart Surgery
Published on: August 12, 2021
2.0K
Development of a Grape Cut Point Detection System Using Multi-Cameras for a Grape-Harvesting Robot
Liangliang Yang1, Tomoki Noguchi1,2, Yohei Hoshino1
1Laboratory of Bio-Mechatronics, Faculty of Engineering, Kitami Institute of Technology, Koentyo 165, Kitami Shi 090-8507, Hokkaido, Japan.
Sensors (Basel, Switzerland)
|January 8, 2025
Summary
This study introduces an AI-powered robot harvester for grapes, utilizing multi-camera systems and object detection to accurately identify stems and determine optimal cutting points, significantly reducing manual labor needs.
Area of Science:
- Agricultural Engineering
- Robotics
- Computer Vision
Background:
- Grape harvesting is labor-intensive, posing challenges for efficiency and cost.
- Automation in agriculture is crucial for addressing labor shortages and improving productivity.
Purpose of the Study:
- To develop an automated robot harvester for vine grapes.
- To create an AI-driven system for precise stem detection and cut point identification.
Main Methods:
- A multi-camera system with a base and hand camera was employed.
- Object detection using You Only Look Once (YOLO) for grape identification.
- Pixel-level semantic segmentation for accurate stem detection and cut point estimation.
Main Results:
- The system achieved high detection accuracy: 98% indoors and 93% outdoors.
- Successful integration of the detection system with a grape-harvesting robot.
- Demonstrated capability for successful outdoor grape harvesting.
Conclusions:
- The proposed AI algorithm and multi-camera system effectively automate grape harvesting.
- The developed robot harvester shows significant potential for practical agricultural applications.
