AI Case Study
PEAT identifies 400 diseases and pests from farmers' uploaded photos on their new mobile app
PEAT has developed a mobile phone app to help farmers identify pests and diseases infecting their plants. The app uses neural networks to train from new user-submitted images and can now identify up to 400 pests and diseases.
From CABI: "CABI, through its Plantwise programme, is partnering with PEAT (Progressive Environmental & Agricultural Technologies) to conduct an 18-month pilot study to assess the benefits of a smartphone app called Plantix. The app gives extension workers and farmers an improved ability to identify economically-important insect pests, diseases and nutrient deficiencies and manage them in order to safeguard crops and livelihoods. German-based company PEAT has developed the Plantix app that identifies plant health problems in the field. The app further offers advice on remedies ranging from conventional control options to preventive measures and biopesticide control. This includes a community feature where agricultural stakeholders can exchange their knowledge. The app works by employing Big Data, Artificial Intelligence (AI) and Deep Neural Networks, or DNNs. The software also has the ability to be multi-lingual and is currently available in Hindi, Telugu and English in India as well as five other languages for use in other countries. During this year, six other local Indian languages will also be launched".
PEAT claims the app can identify up to 400 diseases and pests and has about 200,000 active users which upload 5,000 pictures a day.
From NVIDIA: "PEAT is using two NVIDIA TITAN X GPUs, in scalable link interface mode, in combination with the CUDA parallel computing platform and programming model to not only train the neural network models behind Plantix, but also to analyze incoming photos, a process known as inference. PEAT can calculate a new neural network in 7-10 days on the GPUs".
According to CABI: "Malvika Chaudhary, Asia Regional Coordinator, Plantwise, said, 'Crop losses are mainly caused by plant diseases, weeds, insect pests and soil-nutrient deficiencies. It is important that we use technologies such as Plantix to diagnose problems quickly so that remedies can be applied to prevent further losses.'... Today, farmers in Germany, Brazil and India use Plantix to upload photos of diseased crops. The images are part of a huge and growing crowdsourced database that is helping farmers to identify, treat and prevent crop diseases".
User uploaded photos of plants