Остання редакція: 2023-11-15
Анотація
Abstract. In this work, algorithms for detecting, localizing and tracking moving objects in a stream of video frames are proposed and modeled. The algorithms are based on nonlinear normalized equivalence models, as a measure of the proximity of the template and the current fragment of the video frame, as well as some types of operations on neighboring frames. The results of modeling the proposed algorithms in Mathcad and Labview are presented. The use of equivalence models, measures and frame difference functions gives good results in recognizing and tracking moving objects.
Ключові слова
Посилання
1. Vladimir G. Krasilenko, Alexander I. Nikolsky, Alexandr V. Zaitsev, Victor M. Voloshin, "Optical pattern recognition algorithms on neural-logic equivalent models and demonstration of their prospects and possible implementations", Proceedings of SPIE Vol. 4387 (SPIE, Bellingham, WA 2001), pp.247-260.
2. Krasilenko, V. G., Recognition algorithms of multilevel images of multicharacter identification objects based on nonlinear equivalent metrices and analysis of experimental data / V. G. Krasilenko, A. I. Nikolsky, Yu. A. Bozniak // Управляющие системы и машины. - 2013. - № 4. - С. 12-19.
3. Vladimir.G. Krasilenko, Modelling and comparative analysis of correlation and mutual alignment equivalence functions, /V.G..Krasilenko, Y.A. Boznyak, G.N. Berozov. Science and learning process: scientific and methodical. Proceedings of VSEI Entrepreneurship University "Ukraine". - Vinnitsa: Vinnitsa Social Economy Institute of University "Ukraine", 2009. - P. 68-70.
4. Krasilenko V.G. Using a multi-port architecture of neural-net associative memory based on the equivalency paradigm for parallel cluster image analysis and self-learning // V.G. Krasilenko, A.A. Lazarev, S.K. Grabovlyak, D.V. Nikitovich. - Proceedings of SPIE Vol. 8662 (SPIE, Bellingham, WA 2013) 86620S
5. Krasilenko V.G. Experimental research of methods for clustering and selecting image fragments using spatial invariant equivalent models // V.G. Krasilenko, S.K. Grabovlyak, D.V. Nikitovich. – Proceedings of SPIE Vol. 9286, 2014
6. Krasilenko V.G. Researching of clustering methods for selecting and grouping similar patches using twodimensional nonlinear space-invariant models and functions of normalized "equivalence" / V.G. Krasilenko, D.V. Nikitovich // VII Ukrainian-Polish scientific SPC “Electronics and information technologies” (ELIT-2015). – Lviv:, 2015. – P. 129-134
7. V. G. Krasilenko, A. I. Nikolskyy, and J. A. Flavitskaya „The Structures of Optical Neural Nets Based on New Matrix_Tensor Equivalently Models (MTEMs) and Results of Modeling” ISSN 1060_992X, Optical Memory and Neural Networks (Information Optics), Vol. 19, No. 1, © Allerton Press, Inc., pp. 31–38, 2010.
8. Krasilenko, V.G., Nikolsky, A.I., Bozniak, Y.A., "Recognition algorithms of multilevel images of multicharacter identification objects based on nonlinear equivalent metrics and analysis of experimental data" in Sensor Fusion: Architectures, Algorithms, and Applications VI, edited by Belur V. Dasarathy, Proceedings of SPIE Vol. 4731 (SPIE, Bellingham, WA 2002) pp. 154-163.
9. Krasilenko Vladimir G. "Design and simulation of a multiport neural network heteroassociative memory for optical pattern recognitions"/Vladimir G. Krasilenko, Alexander A. Lazarev, Svetlana K Grabovlyak // Proceedings of SPIE - Vol. 8398.
10. Vladimir G. Krasilenko, Aleksandr I. Nikolskyy, Alexander A. Lazarev, "Modeling optical pattern recognition algorithms for object tracking based on nonlinear equivalent models and subtraction of frames", in MIPPR 2015: Pattern Recognition and Computer Vision, Tianxu Zhang; Jianguo Liu, Editors, Proceedings of SPIE Vol. 9813 (SPIE, Bellingham, WA 2015), 981302.
11. Vladimir G. Krasilenko, Alexander A. Lazarev, Diana V. Nikitovich, "Modeling and possible implementation of self-learning equivalence-convolutional neural structures for auto-encoding-decoding and clusterization of images", Proc.SPIE 10453, Third International Conference on Applications of Optics and Photonics, 104532N (22 August 2017); doi:10.1117/12.2276313; http://dx.doi.org/10.1117/12.2276313