Miary i mapy kinetyczne do automatycznej analizy ruchu w systemach CCTV

• Autorzy: Parzych M. , Marciniak T. , Dąbrowski A.

Identyfikatory

DOI

10.15199/13.2018.4.4

Warianty tytułu

EN

Kinetic measures and maps for automatic motion analysis in CCTV systems

• Języki publikacji: PL

Abstrakty

PL

W artykule przedstawiono metodę obliczania miar ruchu dla inteligentnej analizy wideo. Obliczane są cztery miary: dwukierunkowa długość-łuku, kierunek-w-przód, kierunek-wstecz oraz krawędzie segmentów-ruchu. Wszystkie miary bazują na mapie przepływu obliczonej przy użyciu algorytmu przepływu optycznego. Segmenty-ruchu wykorzystują pola typu finite-time Lyapunov exponent (FTLE). Przeprowadzono eksperymentalne testy dla nagrań wideo z ruchem ludzi. Sprawdzono także czasochłonność obliczeń poszczególnych etapów proponowanego rozwiązania.

EN

This paper presents a calculation of the motion measures for intelligent video analysis. Four measures are proposed: bidirectional length, forward direction, backward direction and motion edges. All measures are based on a flow map obtained with the use of the optical flow algorithm. Motion edges utilizes finite-time Lyapunov exponent (FTLE) fields. Experimental tests were performed for video recordings with the people movement. The computation times of particular steps of the proposed solution were also checked.

Słowa kluczowe

PL

analiza ruchu; ekstrakcja cech; FTLE; LCS; CCTV; biblioteka OpenCV

EN

motion analysis; feature extraction; finite-time Lyapunov exponent fiields; Lagrangian coherent structures; CCTV; OpenCV library

• Wydawca: Wydawnictwo SIGMA-NOT
• Czasopismo: Elektronika : konstrukcje, technologie, zastosowania
• Rocznik: 2018
• Tom: Vol. 59, nr 4
• Strony: 19--23
• Opis fizyczny: Bibliogr. 20 poz., rys., wykr.

Twórcy

autor

Parzych M.

• Poznan University of Technology, Institute of Automation and Robotics Division of Signal Processing and Electronics Systems Poznań, Poland

autor

Marciniak T.

• Poznan University of Technology, Institute of Automation and Robotics Division of Signal Processing and Electronics Systems Poznań, Poland

autor

Dąbrowski A.

• Poznan University of Technology, Institute of Automation and Robotics Division of Signal Processing and Electronics Systems Poznań, Poland

Bibliografia

• [1] Acar Esra, Senst Tobias, Kuhn Aleksander, Keller Iva, Theisel Holger, Albayrak Sahin, Sikora Thomas, 2012, Human action recognition using lagrangian descriptors, International Workshop on Multimedia Signal Processing: 360-365.
• [2] Ali Saad, Shah Mubarak, 2007, A lagrangian particle dynamics approach for crowd flow segmentation and stability analysis, Computer Vision and Pattern Recognition: 1-6.
• [3] BeagleBone Black, dokumentacja techniczna, http://www.adafruit. com/datasheets/BBB_SRM.pdf.
• [4] Blackfin Low Power Imaging Platform, broszura informacyjna, http://http://www.analog.com/media/en/news-marketing-collateral/product-highlight/BLIP-Brochure.pdf.
• [5] Ferryman J. i Shahrokni A, 2009, PETS2009: Dataset and Challenge, Performance Evaluation of Tracking and Surveillance (PETS-Winter): 1-6.
• [6] Garth Christoph, Li Gou-Shi, Tricoche Xavier, Hansen Charles, Hagen Hans, 2009, Visualization of coherent structures in transient 2d flows, Topology-based Methods in Visualization II: 1-13.
• [7] H aller George, 2000, Finding Finite-Time Invariant Manifolds in Two Dimensional Velocity Data, Chaos 10(1): 99-108.
• [8] H aller George, 2001, Langrangian structures and the rate of strain in a partition of two-dimensional turbulence, Physic of Fluids 13(11).
• [9] Kuhn Alexander, Senst Tobias, Keller Ivo, Sikora Thomas, Theisel Holger, 2012, A lagrangian framework for video analytics, Workshop on Multimedia Signal Processing: 387-392.
• [10] Odroid C1, dokumentacja techniczna, http://odroid.com/dokuwiki/doku.php?id=en:cl_hardware.
• [11] Peacock Thomas, Dabiri John, 2010, Introduction to focus issue: Lagrangian Coherent Structures, Chaos 20(1): 017501.
• [12] Raspberry Pie, dokumentacja techniczna, http://elinux.org/RPi_Hardware.
• [13] Sadlo Filip, Bachthaler Sven, Dachsbacher Carsten, Weiskopf Daniel, 2012, Space-time visualization of dynamics in lagrangian coherent structures of time-dependent 2d vector fields, International Conference of Information Visualization Theory and Applications: 573-583.
• [14] Sanderson Allen R., 2014, An alternative formulation of Lyapunov exponents for computing Lagrangian coherent structures, IEEE Pacific Visualization Symposium: 277-280.
• [15] Senst Tobias, Eiselein Volker, Sikora Thomas, 2015, A Local Feature based on Lagrangian Measures for Violet Video Classification, International Conference on Imaging for Crime Prevention and Detection: 1-6.
• [16] Senst Tobias, Kuhn Aleksander, Theisel Holger, Sikora Thomas, 2012, Detecting people carrying objects utylizing lagrangian dynamics, International Conference on Advanced Video and Signal-Based Surveillance: 398-403.
• [17] Shadden Shawn C., Lekien Francois, Marsden Jerrold E, 2005, Definition and properties of Lagrangian coherent structures from finite time Lyapunov exponent in two dimensional aperiodic flows, Physica D 147: 352-370.
• [18] Soori Umair and Mohammad Rizal Arsad, 2009, Underwater crowd flow detection using lagrangian dynamics, Indian Journal of GeoMarine Sciences 38(3): 359-364.
• [19] Suzuki Satoshi, Abe Keiichi, 1985, Topological Structural Analysis of Digitized Binary Images by Border Following, Computer Vision, Graphics and Image Processing 30: 32-46.
• [20] Zach Christopher, Pock Thomas, Bischof Horst, 2007, A Duality Based Approach for Realtime TV-L1 Optical Flow, DAGM Conference on Pattern Recognition: 214-223.

Uwagi

1. Praca sfinansowana ze środków DS w 2018.

2. Opracowanie rekordu w ramach umowy 509/P-DUN/2018 ze środków MNiSW przeznaczonych na działalność upowszechniającą naukę (2018).