In a promising fusion of agriculture and artificial intelligence, researchers at Aberystwyth University are developing a cutting-edge smartphone app designed to detect potato blight before it becomes visible to the human eye.
The disease in question, late blight, is notorious for its rapid spread and devastating impact on potato crops. Responsible for an estimated 20% loss in global potato yields and economic damage exceeding £3.5 billion annually, the fungal disease remains a persistent threat to food security and farmer livelihoods worldwide.
Smart Detection to Save Crops and Costs
The new app is part of the DeepDetect project, which uses machine learning and computer vision to analyse images of potato leaves, identifying signs of infection long before they’re visible. Farmers will be able to simply take a picture with their smartphones and receive real-time, location-specific diagnoses.
“By integrating farmer feedback from the outset, we will ensure this technology meets real-world agricultural needs,” said Edore Akpokodje, a computer science lecturer leading the project at Aberystwyth University.
The tool aims not only to reduce crop losses, but also to cut down the use of harmful pesticides, a major concern in modern farming. Welsh farmers alone spend up to £5.27 million annually on preventive fungicide spraying, a cost the AI app could dramatically reduce.
Potatoes: A Vital Crop Under Threat
Globally, potatoes rank as the fourth most important staple crop, critical to the diets and economies of many countries. In Wales alone, over 17,000 hectares are dedicated to potato cultivation. A disease outbreak not only hits individual farmers but disrupts entire food supply chains.
“Potato blight is not just a farming issue, it’s a food security issue,” said Aiswarya Girija, a researcher at the Institute of Biological, Environmental and Rural Sciences.
The research team is currently compiling a dataset of healthy and infected potato leaves to train and refine the AI model. Once complete, the prototype app will be field-tested with local farmers. The goal is to eventually roll out a national early warning system, with potential to expand to other crops and regions in the future.
Looking Ahead: Smarter, Greener Farming
This innovation represents a broader movement in agri-tech to develop smarter, more sustainable farming tools. As the global population grows and climate challenges intensify, early detection systems like this one could play a vital role in securing crop yields while reducing environmental impact.
“If we can help farmers act early and with precision, we not only save crops, we also reduce unnecessary chemical use and protect ecosystems,” Akpokodje added.
The team hopes that, with additional support and feedback from the agricultural community, their AI-powered tool can set a new standard in disease detection, making farming more resilient, efficient, and eco-friendly.
Read also: Cocoa Crisis Exposed: Why Africa’s Chocolate Gold is Failing Farmers
