AI Tool Automates Plant Fruit Measuring to Breed Better Crops
23 Aug 2025 | 3 minute read
Researchers at Aberystwyth University in Wales have developed an AI-based tool that automatically measures plant seeds and seed pods to improve crop breeding. This approach addresses the inefficiencies of traditional methods, which are laborious, slow, and prone to human error.
The AI tool analyses images to accurately identify seed pods and measure key traits such as pod length, width, area, and volume. These characteristics directly influence crop yield and profitability, making precise measurement vital for breeding programs.
Linking Traits to Genetics
The study connects these physical traits to specific genetic regions responsible for pod shape and size. This helps scientists identify genes that control plant development and growth, providing targets to enhance yield, shape, and resilience through breeding.
The AI tool has been tested on various crops, including oilseed rape, cabbages, and cereals like oats, barley, and wheat. Its adaptability suggests potential applications across a wide range of plant species.
Scaling Phenotyping with AI
Kieran Atkins, PhD researcher and project lead from IBERS at Aberystwyth University, said the AI system gathered data on over 300,000 individual fruits. This volume demonstrates how deep learning can handle large-scale phenotyping efficiently, eliminating technical and time constraints.
By simplifying access to phenotyping, the tool allows more researchers to investigate plant traits at a scale previously impractical, opening new avenues for discovery and innovation.
Transforming Genetic Analysis and Breeding
Professor John Doonan, Director of the National Plant Phenomics Centre, emphasized that deep learning AI delivers the data quality and accuracy required for genetic studies and crop improvement. Advanced imaging combined with AI bridges plant morphology and genetic function more effectively.
Originally developed for a small model plant, the tool has proven effective on brassica crops, marking progress toward scalable, data-rich phenotyping. This supports predictive approaches in crop breeding by accelerating research and improving trait selection.
The team has made their MorphPod tool available online, encouraging researchers worldwide to replicate or adapt the system for other plant species.
For those interested in AI applications in biological research or plant science, exploring courses on AI tools for scientific analysis can provide practical insights into these emerging technologies.
Your membership also unlocks:
