Scientists develop a system for monitoring Altai's biodiversity with Yandex Cloud
With the help of ML algorithms, scientists can determine the spread of endangered tree and grass species, and farmers can prevent crop losses.
A team of scientists from Altai State University, together with Yandex Cloud, created a plant monitoring system based on artificial intelligence.
ML algorithms predict the spread of plants in the territory specified in the system. In addition to scientific monitoring, algorithms can help solve applied problems of agricultural producers. About 17% of the grain harvest in Russia is lost annually due to the crop contamination. With the help of the new system, farmers will be able to more accurately predict the spread of pest plants of each individual species. This will allow you to determine the best locations and the exact area for crops, as well as plan harvesting and tillage methods.
The system deployed on the Yandex Cloud cloud platform predicts the spread of a plant based on several data types — information about the current points of presence of this species, the local climate, humidity, and soil type features. The user can enter the Latin name of a plant into the system, upload a dataset, and, after processing the request, receive graphs and detailed forecast maps. The generated heat map allows you to estimate exactly where the plant will be found in the future.
The tool is currently operating in an experimental mode and is being tested on plant species in the Altai region. In the future, the system will make forecasts in other regions of Russia. Also, together with Yandex Cloud, Altai Federal University scientists plan to use computer vision to improve the tool. The user will not be able to enter the type of plant into the forecast system, but will be able to take a picture of the plant using a smartphone camera and upload a photo to the cloud. Using the Yandex Vision image recognition service, the system will automatically detect each species.
“Opportunities at the intersection of life sciences and data sciences open the door for us to model complex processes and plan outcomes. It’s worth noting that the capabilities of the monitoring system are very broad and go far beyond the limits of scientific curiosity. Today, the development team and I see a whole range of services in the future for the agrobioindustry, health, and environmental protection markets behind this virtual biolab.”
“Scientists are increasingly using cloud technologies in their research activities. There are several reasons: using the cloud speeds up project development, allows you to increase computing power with just a click, and maintain a high level of data security. In addition, the cloud remains a convenient platform where you can easily access ready-made managed services needed for experimentation.”
