AN INVESTIGATION OF AGGREGATE CHANNEL FEATURES OBJECT DETECTOR FOR UAS APPLICATION
Keywords:UAS, artificial intelligence, Aggregate Channel Features, object detection, video-stream
Purpose: The paper is aimed to point out an artificial intelligence as a key priority for research and development issues. Consideration of existing methods of object detection is one of the important tasks of the paper. The represented research results are aimed to investigate a problem of moving object detection by visual sensor data for Unmanned Aerial System application. Methods: Represented approach is grounded on probabilistic and statistical methods of data processing, in particular on Aggregate Channel Features approach usage for object detection. Results: An Aggregate Channel Features approach for object detection by video-stream at different scenarios has been practically investigated. Results of experimental investigation of ACF usage for moving vehicles detection, such as cars and trams, indicate good performance characteristics. Also, a dependence between training time of detector and amount of object positive instances has been investigated for particular case. Discussion: Multiple advantages of Aggregate Channel Features object detector such as universality, simplicity of realization and good compromise between computation time and detection accuracy allow to use it in the tasks of people, vehicles, artificial and natural objects detection in UAS application. Represented results can be implemented in Unmanned Aerial Systems for searching and tracking of movable objects.
European Commission (2018) Communication from the Commission to the European Parliament, the European Council, the Council, the European economic and social committee and the committee of the regions. Artificial Intelligence for Europe, Brussels, 19 p.
IATA (2018) AI in aviation. Exploring the fundamentals, threats and opportunities of artificial intelligence (AI) in the aviation industry. White Paper, 20 p.
European Commission (2018) Declaration of cooperation on Artificial Intelligence, Brussels. Available at: https://ec.europa.eu/jrc/communities/sites/jrccties/files/2018aideclarationatdigitaldaydocxpdf.pdf
Viola P., Jones M. (2001) Rapid object detection using a boosted cascade of simple features. IEEE Conference on Computer Vision and Pattern Recognition, 9 p.
Freund, Y. and Schapire, R.E. (1997) A decision-theoretic generalization of on-line learning and an application to boosting. Journal of computer and system sciences, 55(1), pp.119-139.
Jones, M.J. and Viola, P., (2001) Robust real-time object detection. In Workshop on statistical and computational theories of vision, Vol. 266, p. 56.
Angelova, A., Krizhevsky, A. and Vanhoucke, V. (2015) Pedestrian detection with a large-field-of-view deep network. In Robotics and Automation (ICRA), IEEE International Conference, pp. 704-711.
Pérez, A., Larrañaga, P. and Inza, I., (2009) Bayesian classifiers based on kernel density estimation: Flexible classifiers. International Journal of Approximate Reasoning, 50(2), p.341.
Kharchenko, V., Kukush, A., Kuzmenko, N. and Ostroumov, I., (2017). Probabilistic Approach to Object Detection and Recognition for Videostream Processing. Proceedings of the NAU, № 2(71), pp.8-14.
Benenson, R., Mathias, M., Timofte, R. and Van Gool, L., (2012) Pedestrian detection at 100 frames per second. In Computer Vision and Pattern Recognition (CVPR), IEEE Conference, pp. 2903-2910.
Dalal, N. and Triggs, B., (2005) Histograms of oriented gradients for human detection. In Computer Vision and Pattern Recognition. CVPR 2005. IEEE Computer Society Conference on Vol. 1, pp. 886-893.
Dollar P., Tu Z., Perona P., Belongie S. (2009) Integral Channel Features. British Machine Vision Conference, 11p.
Dollár, P., Belongie, S.J. and Perona, P., (2010) The fastest pedestrian detector in the west. In Bmvc, Vol. 2, No. 3, p. 7.
Dollár, P., Appel, R., Belongie, S. and Perona, P., (2014) Fast feature pyramids for object detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 36(8), pp.1532-1545.