A Uav-based multisensor framework for legal industrial Cannabis monitoring and open-access dataset development.

Abstract

Industrial hemp cultivation is expanding and requires reliable monitoring for legal compliance and agricultural management. This paper presents a standardized UAV-based multisensor framework designed for Cannabis sativa L. It integrates RGB, multispectral, and thermal imaging as core modules, with hyperspectral and LiDAR as optional extensions. The framework sets protocols for sensor integration, flight planning, field measurements, and annotation, ensuring datasets that meet EU altitude limits (≤120 m AGL). Multi-altitude and multi-time-of-day acquisitions are proposed to capture spatial and diurnal variability. These data improve model robustness for phenotyping, stress detection, and THC compliance verification. Potential applications include precision agriculture, breeding, regulatory monitoring, environmental assessment, and illicit crop detection. Open-access datasets generated through this framework will support reproducibility, machine learning development, and collaboration among researchers, farmers, and regulators.

Plant phenotyping relevance

UAVマルチセンサーフレームワークの設計、取得プロトコル、アノテーション、オープンデータセット開発が中心で、植物表現型やストレスを測定する方法論的貢献が明確です。

Code and data availability

This is a perspective/framework paper proposing a UAV-based multisensor monitoring protocol for industrial hemp. No dataset, imagery, code, or models are yet available: the authors state the dataset will be collected starting in 2026 and made public only 'once field access is permitted', contingent on regulatory approv

No evidence-backed public reproduction asset is currently recorded.