Researchers at the Indian Institute of Technology (IIT) Mandi have developed a computational model using Artificial Intelligence (AI) for automated disease detection in potato crops using photographs of their leaves. This research, conducted in collaboration with the Central Potato Research Institute, Shimla, uses AI techniques to highlight the diseased portions of the leaf and has also been published in Plant Phenomics.
Scientists at IIT Mandi developed a computational tool for detecting blight in potato leaf images. The model makes use of an AI tool called mask-region-based convolutional neural network architecture to identify diseased portions of the leaf amid a complex background of plant and soil elements. Further efforts are being made to convert the developed tool into a smartphone application for more practical use.
In the history of the world, potatoes were the cause of the great Irish famine that killed more than a million people and destroyed the Irish language in the mid-nineteenth century. What is the reason? The potato blight.
Srikant Srinivasan, Associate Professor, School of Computing and Electrical Engineering, IIT Mandi, said, ‘The blight is a common disease of the potato plant, that starts as uneven light green lesions near the tip and the margins of the leaf and then spreads into large brown to purplish-black necrotic patches that eventually leads to rotting of the plant. If left undetected and unchecked, blight could destroy the entire crop within a week under conducive conditions.’
‘In India, as in most developing countries, the detection of blight is performed manually by trained personnel who visually inspect potato foliage,’ he said adding, ‘This process is tedious and impractical, especially in remote areas, because it requires the expertise of an expert who may not be physically accessible.’
A research scholar at IIT Mandi explained that automatic disease detection can help in this regard and given the proliferation of smartphones across the country, the smartphone could be a useful tool in this regard. ‘The advanced HD cameras, better computing power and strong communication capabilities offered by smartphones provide a promising platform for automated detection of pathogens in crops, which will save time and help in the timely management of disease outbreaks,’ he said.
Researchers collected data on healthy and diseased leaves from Punjab, Uttar Pradesh and Himachal Pradesh to develop a robust model.
According to Srinivasan, ‘It was imperative that the model developed is portable across the nation. An analysis of the detection performance indicates an overall precision of 98 percent on leaf images in field environments’
Even though potato is not a staple food in most regions of the world, it is a cash crop and failure to cultivate it can have disastrous consequences, particularly for farmers with smallholdings.
To prevent financial catastrophe for the farmer and the country, early detection of blight is crucial.
In light of this success, Srinivasan currently is shrinking the model to be easily hosted on a smartphone as an application. This way, when the farmer photographs the leaf which appears sickly, the application confirms in real-time if the leaf is infected or not.
‘With this timely information, the farmer would know when to spray the field, saving his produce and reducing costs associated with unnecessary use of fungicides,’ said Srinivasan. ‘The model is being further refined as more states are included.’
The app will be distributed as part of the FarmerZone app that potato farmers will be able to download for free.
The research team also includes Shyam K Masakapalli from IIT Mandi along with Joe Johnson and Geetanjali Sharma; Vijay Kumar Dua, Sanjeev Sharma and Jagdev Sharma from the Central Potato Research Institute in Shimla.