The Sigma-Pro company, an NSU industrial partner, turned to engineers at NSU鈥檚 Laboratory of Applied Digital Technologies with a request to develop software for a new class of problems in experimental studies in the field of fluid and gas mechanics. For example, to develop anti-icing agents for aircraft, to study aerosol drug delivery systems, to create spray cooling systems, to apply paintwork, etc. Interns Ksenia Borisova and Timur Garipov, students of the Engineering School that belongs to NSU`s , were involved in automating the process of object recognition using machine vision methods.
Rustam Mullyajanov, Head of the Laboratory, Doctor of Physical and Mathematical Sciences, talked about their work,
Our Laboratory is engaged in the development and implementation of algorithms for solving applied problems related to mechanics, hydrodynamics, heat and mass transfer, bioengineering, and natural language processing by combining the results of supercomputer modeling, experiments, and modern machine learning methods. Currently, 芦Sigma-Pro禄 is developing its own software for experimental research and testing in the field of fluid and gas mechanics. Students from the Engineering School are regularly involved in our projects, and this case was no exception.
Under the guidance of Laboratory researcher Ivan Plokhikh, the students learned to use neural networks to recognize drops of water spray from a nozzle in a supersonic flow. The data obtained was processed to collect information about the centers of mass, sizes, velocities, and movements of droplets in paired images.
Timur Garipov described his experience,
My role in the project was to mark up the data to train the model, as well as to develop and modify a neural network for segmenting drops in the image. Initially, the customer provided data in the form of a set of pairs of consecutive images. Then, together with Ivan, I was engaged in the selection of architecture and modification of the solution to obtain a suitable result. I enjoyed working with 芦Sigma-Pro禄. I think it was an important and valuable experience to work with an industry representative while I was still a student. I am extremely glad that I was able to apply my knowledge to address a practical problem.
The developed algorithm has been successfully implemented in the ActualFlow software for the POLIS complex hardware base that was previously launched jointly with the S. S. Kutateladze Institute of Thermal Physics SB RAS. Research organizations (research institutes, universities) and scientific divisions of commercial companies are already requesting the software.
