Innovations: AI will detect potato blight symptoms before they appear
Welsh scientists are developing an innovative mobile app for the early detection of potato blight. Specifically, AI will detect blight symptoms before they are visible to the naked eye.
Potato blight, caused by Phytophthora infestans, is one of the most dangerous diseases in potato crops, responsible for 20% of crop losses worldwide. Its control significantly increases production costs, as it requires the use of fungicides.
The key to effective crop protection lies in detecting potato blight symptoms as early as possible. To this end, scientists at Aberystwyth University are implementing the DeepDetect project. This project seeks more precise and sustainable disease management. Its main objective is to provide farmers with real-time, location-based disease forecasts directly on their smartphones.
Protection against late blight in potatoes significantly increases production costs because it requires the use of fungicides.
Artificial intelligence algorithms for plant protection
The DeepDetect project is based on advanced machine learning algorithms that analyze images and environmental data to predict infection risk. The system will use thousands of images of leaves, both healthy and infected, to identify early signs of disease, even imperceptible to the naked eye. Based on this, it will provide accurate disease predictions, taking into account the location of a specific farm.
“Our goal is to provide farmers with a tool that is not only scientifically sound, but also practical and easy to use, delivering instant, location-specific disease forecasts directly to their phones,” says Dr. Edore Akpokodje, a computer scientist at Aberystwyth University and project leader.
AI will detect potato blight symptoms before they are visible to the grower.
Traditional methods for detecting potato blight rely primarily on crop monitoring and visual assessment. This, in turn, is not only labor-intensive but also error-prone. Meanwhile, a new AI-based application will make it possible to detect the first symptoms of the disease before they are visible to the human eye. This will enable immediate and targeted intervention.
Therefore, AI technology will allow for better adjustment of treatment timing to actual phytosanitary risks. This will reduce the overuse of fungicides and minimize environmental impact.
Prospects for implementing an innovative application
The DeepDetect project has received funding from the Smart Flexible Innovation Support (SFIS) program, funded by the Welsh Government. Work is currently underway to develop a prototype model, which will then be trained on large image sets, including healthy potato leaves and those with blight symptoms. Consultations and field testing with farmers will also be conducted to ensure the tool’s practical usefulness.
Ultimately, the system aims to serve as a national early warning system for potato blight. Subsequent stages will allow for its adaptation to monitor other plant diseases and implementation in other regions and crops.
Yes, by better matching treatment timing to actual risk, technology can help reduce overuse of fungicides and minimize environmental pressure.
Fuente:
