< br />
NOVOSIBIRSK, January 11 Scientists at Novosibirsk State University (NSU) have developed a unique and universal service for processing microscopy images, which includes training a neural network, the university reported on Thursday.
«»The universal Telegram service for recognizing numerous homogeneous objects of various types includes training a neural network by the user using a marked area from an image he has uploaded. This service can be used not only in microscopy and materials science, but also in other areas of science and technology,» — the message says.
The university noted that NSU scientists presented the recognition service they developed for numerous homogeneous objects of various types in the article “DLgram cloud service for deep analysis of microscopic images,” which was published in the prestigious journal Microscopy Research and Technique. The work was supported by a grant from the Russian Science Foundation.
“We presented the first version of this service back in 2021. Then, thanks to the support of a grant from the Russian Science Foundation, we improved it: we expanded the functionality, introduced new capabilities for analyzing recognized objects. Now our Telegram service is quite famous among microscopists and has more than 300 users from Russia and abroad milestone,” said Andrey Matveev, head of the laboratory of deep machine learning in physical methods at the Institute of Intelligent Robotics at NSU.
He explained that no special skills are required to work with the service. A neural network frees a person from performing routine processes associated with counting and characterizing particles or objects in an image — determining their quantity and concentration, as well as maximum, minimum and average sizes.
“The service operates in cloud mode, that is, the Cascade Mask-RCNN neural network used does not work on the user’s computer, but on the graphic server of the Institute of Intelligent Robotics of NSU,” the scientist noted.
The new service works with various photographic images — pictures from microscopes, cameras and mobile phones, and the neural network recognizes not only particles of substances, but also macro-objects.
«We are actively developing the creation of such digital assistants not only in the field of microscopy. Staff and students of the institute are working on the creation of automatic recognition systems for various objects in industry, as well as for the urban environment as part of the federal project for the development of Artificial Intelligence Centers,» — said the director of the Institute of Intelligent Robotics of NSU Alexey Okunev.