How the System Works
We train and test our systems end-to-end on our large Pests and diseases Data set, which contains challenging images with diseases and pests, including several inter and extra class variations, such as infection status and location of the plant where it has been taken from. Experimental results show that our proposed system can effectively recognize seven different types of diseases and pests, with the ability to deal with complex scenarios from a plant’s surrounding area. Diseases reduce yields when we need them to be high, as the global population grows and we need to feed more people than ever. Reducing yield loss to diseases like Cercospora leaf spot in Maize, Rice Tungro disease, cassava mosaic disease and many others will allow us to grow more food, more efficiently and on less land, benefiting the agricultural industry, the public and the environment.
A farmer can use a drone to detect any anomalies, upload a photo of infected crops and send it offline to the farmer’s mobile app. The mobile app does the processing, it broadcasts that disease to the nearby farmers within that community with all descriptions about that disease, the contact and location of the specialized agronomist to work on that crop and give advice in real-time. Besides giving a diagnosis and steps to mitigate the disease, the app also provides information on preventing the disease in the next cropping season.