METHODOLOGY FOR PREDICTING FLAX FIBER AND SEED YIELDS USING UAV PHOTOGRAPHY AND AUTOMATED PHOTOANALYSIS

Abstract

Accurate forecasting of flax (Linum usitatissimum) fiber and seed yields is essential for optimizing harvest operations
and post-harvest processing. This paper presents a methodology that integrates high-resolution UAV (drone) imagery with field
observations of crop maturity and growth quality to predict final fiber and seed yields. A multi-rotor UAV equipped with
multispectral and RGB sensors captured images at key growth stages. These images were automatically processed using a
custom image-analysis module to compute vegetation indices (e.g., NDVI) and structural plant traits. The proposed method
provides timely and precise yield assessments, supporting efficient flax production and sustainable utilization of agricultural
biomass.