Researchers are developing a new app which will use artificial intelligence to warn farmers of a fungal disease which can devastate potato crops.
Welsh scientists say the app will allow farmers to detect late blight using their phone before it becomes visible to the human eye.
The disease is responsible for 20% of potato crop losses and £3.5bn in economic losses worldwide, the research team at Aberystwyth University said.
Early diagnosis would boost productivity and reduce costs for farmers, as well as decrease their reliance on environmentally harmful pesticides, they added.
The DeepDetect project will use machine learning to deliver accurate, location-specific disease diagnoses to farmers on their smartphones.
“By integrating farmer feedback from the outset, we will ensure that this technology is grounded in real-world needs and challenges,” said Edore Akpokodje, computer science lecturer at Aberystwyth University.
The technology also has the potential for wider application across other crops, he said.
The project aims to reduce the environmental and financial burden of preventive spraying, which the researchers said currently cost Welsh farmers up to £5.27m annually.
Potatoes are a vital crop globally and in Wales over 17,000 hectares are dedicated to potato farming.
The team plans to create an AI-powered prototype using image datasets of healthy and diseased potato leaves.
It will then refine the model and eventually the team hopes to create a national early warning system for potato blight, with potential to expand the technology to other crops and regions in the future.
“Potatoes are the fourth most important staple crop globally, and optimal production is essential for a growing global population,” Aiswarya Girija from the Institute of Biological, Environmental and Rural Sciences at Aberystwyth University said.
“Potato blight is therefore not just a farming issue – it’s a food security issue.”
