Researchers at the University of Tokyo and Kubota Corporation have developed a drone-based system that predicts underground potato yield before harvest by combining drone imagery, machine learning, and a mathematical growth model. Field trials in 2023 and 2024 achieved high accuracy (0.8+ correlation for biomass, 0.7+ for yield), offering farmers a non-destructive alternative to traditional sampling methods and supporting better harvest timing decisions.
Summaries like this, in your inbox every morning.
Sign up free →What happened
Researchers at the University of Tokyo and Kubota Corporation developed a drone-based system that estimates underground potato yield before harvest. The method combines drone photography (using RGB and multispectral cameras), machine learning trained on the relationship between plant features and actual biomass, and a Gompertz growth curve model. Two-year field trials in 2023 and 2024 achieved a correlation coefficient of 0.8 or higher for tuber biomass estimation and 0.7 or higher for yield prediction.
Why it matters
Traditionally, assessing potato yield during growing season has relied on destructive sampling (physically digging to measure). This non-destructive approach can capture spatial variation across a field and support pre-harvest yield forecasting and optimization of harvest timing—practical improvements for growers. The development reflects precision agriculture use cases that Tokyo-based Market Research Center forecasts will drive Japan's agriculture drone market to reach $357.8 million（約570億円） by 2034.
What to watch
The research team says the growth-curve approach is expected to support optimization of cultivation management, including suggesting optimal harvest timing. The work was carried out under the joint Kubota Todai Lab project.
No comments yet. Be the first to share your thoughts!
Log in to join the discussion
Get curated AI news from 200+ sources delivered daily to your inbox. Free to use.
Get Started FreeFree · takes 30 seconds · unsubscribe anytime
1 minute a day. The AI essentials.
200+ sources · Email / LINE / Slack
