An automated disease detection program has been developed to spot diseases in potato leaves, a July 12 news release from the Indian Institute of Technology (IIT) Mandi says.
Scientists from IIT Mandi, led by Srikant Srinivasan, an associate professor in the school of computing and electrical engineering, in collaboration with the Central Potato Research Institute, Shimla, were able to use artificial intelligence (AI) to find diseased parts on leaves.
Blight is a common potato disease which starts as uneven light green lesions near the tip and margins of the leaf, the release says. It then turns into large brown to purplish/black necrotic patches eventually causing the plant to rot. If left undetected and unchecked, blight can destroy an entire crop within a week in the right conditions.
“In India, as with most developing countries, the detection and identification of blight are performed manually by trained personnel who scout the field and visually inspect potato foliage,” Srinivasan explains in the release. The process is tedious and impractical.
“Automated disease detection can help in this regard and given the extensive proliferation of the mobile phones across the country, the smartphone could be a useful tool in this regard,” Joe Johnson, research scholar with IIT Mandi, says in the release.
Advanced HD cameras, better computing power and communication avenues available on smartphones, are making for a promising platform for automated disease detection in crops. This can allow for timely disease management and stop outbreaks, the release says.
The tool developed by IIT Mandi scientists detects blight in potato leaves by taking photos. An AI tool called mask region-based convolutional neural network architecture is used to accurately highlight diseased portions of leaves. It was able to detect 98 per cent of leaf images in field environments, the release notes.
The research team is now working on making it a smaller megabyte size so it can become a smartphone application. This will allow growers to take photos of unhealthy looking leaves and get real-time answers on if it’s infected or note. It will be included as part of the FarmerZone app which will be available for free.
The research was recently published in the journal Plant Phenomics.