Affiliation:
1. Belarusian State University
2. Nanjing Research Institute of Electronics Engineering
3. Belarusian State University; United Institute of Informatics Problems of the National Academy of Sciences of Belarus
Abstract
Nowadays, homogeneous objects clusters motion is one of the most important and rapidly developing computer
vision and machine learning application. In this paper, we consider the crowd motion patterns determination
by using motion maps that we calculate with FlowNet, a neural network examining motion of objects in a video
sequence. This approach allows us to get information on the crowd direction and velocity with relation to other
objects of scene, which plays the key role in behavior analysis and security establishment. Besides, we consider
methods for preliminary video sequence processing, including frame combination, to estimate motion maps more
precisely and improve the effectiveness of the dynamic scenes analysis.
Reference13 articles.
1. Ascione, F., Caserta, S., Perris, R., & Guidoa, S. (2014). Investigation of Cell Dynamics in vitro by Time Lapse Microscopy and Image Analysis. Chemical Engineering, (38), 517–522. DOI: 10.3303/CET1438087.
2. Chen C., Ye S., Chen H., Nedzvedz, O. V., & Ablameyko, S. V. (2017). Integral Optical Flow and its Application for Monitoring Dynamic Objects from a Video Sequence. J. of Applied Spectroscopy, (84), 120–128. DOI: 10.1007/s10812-017-0437-z.
3. Sholtanyuk, S. (2023). Crowd Abnormal Behaviour Patterns: Survey and Detection. Central European Researchers Journal, 9(1), 48-58. https://ceres-journal.eu/download.php?file=2023_01_7.pdf.
4. Altowairqi, S., Luo, S., & Greer, P. (2023). A Review of the Recent Progress on Crowd Anomaly Detection. Intern. J. of Advanced Computer Science and Applications, 14(4), 659–669. DOI: 10.14569/IJACSA.2023.0140472.
5. Choudhry, N., Abawajy, J., Huda, S., & Rao, I. (2023). A Comprehensive Survey of Machine Learning Methods for Surveillance Videos Anomaly Detection. IEEE Access, (11), 114680–114713. DOI: 10.1109/ACCESS.2023.3321800.