Intelligent analysis of video data in system for monitoring compliance with industrial safety rules

The use of intelligent cameras and sensors, in combination with the human operator in video analytics systems, from which most of the analytical and visual load has been removed, allows you to increase the efficiency of video surveillance and, as a result, increase the safety and productivity of work in production as a whole. Analysis of the existing data processing methods in the video surveillance systems of industrial facility showed that the use of a non-contact method for analyzing person’s posture and actions in the camera’s field of vision is rare, but it can be critical in certain situations (person in overalls is in the camera’s field of view, but the system is on him does not respond, because he is not in the forbidden zone). The improvement of algorithms for the intellectual analysis of video data in the system for monitoring compliance with industrial safety rules (analysis of the type of dynamics and control "friend or foe") using neural network processing technologies is considered. Effectiveness evaluation of algorithms for analyzing full-scale video data software implementation showed the correctness of classification in 97% of cases. Effectiveness evaluation of the 5 subjects into two classes of “own” and “alien” classification was carried out by cross-validation and showed an accuracy of 99% on the test sample.

1. Zhukova P.N., Nasonova V.A., Prokopenko The transport infrastructure safety by the use of video surveillance and video analytics. Problems of law-enforcement activity. 2015;4:91- 96. (In Russ)

2. Zabashta A.Ju., Skorikova S.A. Video analytics functions, analysis of video analytics systems architectures. Rostov Scientific Journal. 2017;7:194-200. (In Russ)

3. Center 2M - Russian information operator. Available at: https://center2m.ru/ (accessed 01.04.2020). (In Russ)

4. ISS – Intelligent Security Systems. Available at: https://iss.ru/ (accessed 01.04.2020). (In Russ)

5. Automated Detection and Alerting in Seconds with Visual Confirmation of Events. Available at: https://intelliviewtech.com/solutions/video-analytics/ (accessed 01.04.2020).

6. Marr D. Vision: An Informational Approach to Studying the Representation and Processing of Visual Images. M.: «Radio i svjaz». 1987:2. (In Russ)

7. Toshev A., Szegedy C. Deeppose: Human pose estimation via deep neural networks. Proceedings of the IEEE conference on computer vision and pattern recognition. 2014:1653-1660.

8. Bulat A., Tzimiropoulos G. Human pose estimation via convolutional part heatmap regression. European Conference on Computer Vision. Springer, Cham. 2016:717-732.

9. Krizhevsky A., Sutskever I., Hinton G. E. Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems. 2012:1097-1105.

10. Ofli F. et al. Berkeley mhad: A comprehensive multimodal human action database. 2013 IEEE Workshop on Applications of Computer Vision (WACV). IEEE. 2013:53-60.

11. Greff K. et al. LSTM: A search space odyssey. IEEE transactions on neural networks and learning systems. 2016;28(10):2222-2232.

12. Cao Z. et al. OpenPose: realtime multi-person 2D pose estimation using Part Affinity Fields. arXiv preprint arXiv:1812.08008. 2018.

13. Buoy T.T.Ch., Fan N.Kh., Spitsyn V.G. Face recognition based on the application of the Viola-Jones method, wavelet transform and principal component analysis. News of Tomsk Polytechnic University. 2012;320(5):54-59. (In Russ)

14. Chuikov A.V., Vulfin A.M., Vasilyev V.I. Neural network system for converting the user biometric characteristics into a cryptographic key. TUSUR report. 2018;21(3):35-41. (In Russ)

15. Chuikov A.V., Vulfin A.M. Gesture recognition system. Vestnik UGATU. 2017;21(3 (77)):113-122. (In Russ)