Developers from the Yandex School of Data Analysis, together with Yandex B2B Tech and the Far Eastern Federal University, have created a neural network capable of speeding up the process of cleaning up the shores of water bodies from debris. The technology is based on the analysis of aerial photographs, and it allows determining the volume, mass, and types of waste on the coasts.
The neural network has already been successfully tested in the South Kamchatka Federal Reserve and in the Arctic. According to company representatives, the neural network helped determine that the coast is most polluted by plastic containers and packaging, as well as waste from industrial fishing.
Cleaning the shore requires a team of 20 volunteers and special equipment. The use of drones and a neural network made it possible to remove five tons of debris four times faster than without the use of technology.
In 2025, it is planned to expand the use of the neural network in other reserves of the Far East and the Arctic. The development code is open source, which will allow it to be used to monitor various types of waste around the world.
Read more on the topic:
Yandex Develops First Reasoning Neural Network in Russia
Updated Search with Neural Network Functions Launched in Yandex
Neural Network for Detecting Rare Pathology Created in Yandex
