Issue |
ITM Web Conf.
Volume 59, 2024
II International Workshop “Hybrid Methods of Modeling and Optimization in Complex Systems” (HMMOCS-II 2023)
|
|
---|---|---|
Article Number | 03003 | |
Number of page(s) | 6 | |
Section | Data Mining, Machine Learning and Patern Recognition | |
DOI | https://doi.org/10.1051/itmconf/20245903003 | |
Published online | 25 January 2024 |
Neural network algorithm for predicting human speed based on computer vision and machine learning
1
Department of Automated Systems of Decision-Making Support, Tambov State Technical University,
Tambov,
392000,
Russian Federation
2
Department of Mechatronics and Technological Measurements, Tambov State Technical University,
Tambov,
392000,
Russian Federation
* Corresponding author: obuhov.art@gmail.com
The problem of increasing the accuracy of predicting human actions is an urgent task for various human-machine systems. The study examines the solution to the problem of predicting human speed using neural network algorithms, computer vision technologies, and machine learning. The formalization and software implementation of a neural network speed prediction algorithm are presented. To solve the problems of determining the current speed and predicting the upcoming positions of the human body depending on the dynamics of its movement, a comparison of various machine learning models was carried out. The RandomForestRegressor algorithm showed the best position prediction accuracy. The best determination of the current speed was demonstrated by dense multilayer neural networks. The experiment revealed that when predicting a person's position at an interval of 0.6 seconds, his speed is determined with an accuracy of more than 90%. The results obtained can be used to implement neural network algorithms for controlling human-machine systems.
© The Authors, published by EDP Sciences, 2024
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.