Authors: Matthew Gazzard, Helen Hicks, Isibor Kennedy Ihianle, Jordan J. Bird, Md Mahmudul Hasan, Pedro Machado

Published on: May 12, 2024

Impact Score: 7.8

Arxiv code: Arxiv:2405.07349

Summary

• What is new: Introduction of a Real-Time Autonomous Black-Grass Classification and Mapping system (RT-ABGCM) using advanced YOLO models for immediate weed detection.
• Why this is important: Blackgrass severely impacts food security by reducing crop yields and increasing cultivation costs, with an additional environmental toll due to herbicide use.
• What the research proposes: The RT-ABGCM system uses AI algorithms to detect blackgrass in real-time and map its density, assisting in precision weed management.
• Results: Showcased effective real-time detection and mapping of blackgrass, enhancing the potential for precise management and reducing herbicide usage.

Technical Details

Technological frameworks used: YOLOv8, YOLO-NAS

Models used: YOLO models accelerated with NVIDIA Jetson Nano

Data used: Two datasets specific to blackgrass detection

Potential Impact

Agricultural sector, specifically companies in crop management, precision agriculture solutions, and agricultural AI technologies

Want to implement this idea in a business?

We have generated a startup concept here: AgriScout AI.