Conteo de personas con un sensor RGBD comercial

Castrillón-Santan, M.; Lorenzo-Navarro, J.; Hernández-Sosa, D.

doi:10.1016/j.riai.2014.05.006

Información del artículo

Resumen

Texto completo

Bibliografía

Descargar PDF

Estadísticas

Resumen

En este trabajo se demuestra que la información de profundidad proporcionada por una cámara RGBD comercial de bajo coste, es una fuente fiable de datos para realizar de forma robusta el conteo automático de personas. La adopción de una configuración de vista cenital reduce la complejidad del problema, al mismo tiempo que permite preservar la privacidad de las personas moni- torizadas. Para llevar a cabo el estudio experimental se han considerado dos técnicas propias del campo de análisis de imágenes 2D trasladadas al contexto de imágenes de profundidad. Las pruebas evaluaron su rendimiento con v¿ıdeos reales sin restricciones de iluminación, incluyendo episodios de iluminación cambiante o muy baja. En este conjunto experimental se realizó la detección, seguimiento y análisis de patrones de comportamiento de las personas que cruzaban el campo de visión. Los resultados obtenidos alcanzan una tasa de acierto próxima al 95%, superando los obtenidos con técnicas actuales basadas exclusivamente en información visual. Estos resultados sugieren la utilidad del uso de información de profundidad en esta tarea particular.

Palabras clave:

Conteo de personas

cámaras de profundidad

detección de eventos

detección de objetos

Abstract

In this paper, we prove that depth information provided by a consumer depth camera is a reliable data source to perform ro- bust people counting. The adoption of a top view configuration reduces the space problem complexity for this task, while pre- serving privacy. Two different background subtraction approaches for color images are transferred to this context and tested in real video to perform detection, tracking, and behavioral pat- terns analysis of subjects crossing the field of view. The results achieved in an experimental setup with real video reported a TPR over 95%, beating robust GMM background subtraction based only on the visual cue. The results suggest the benefits of the depth cue for this particular task.

Keywords:

People counting

Consumer depth cameras

Event detection

Object detection

Texto completo

Referencias no citadas

Albiol et al., 2012, Albiol et al., 2001, Albiol and Silla, 2010, Andriluka et al., 2008, Antic et al., 2010, Barandiaran et al., 2008, Barbosa et al., 2012, Bellotto and Hu, 2009, Beymer, 2000, Blanco et al., 2003, Bozzoli et al., 2007, Brutzer et al., 2011, Camplani et al., 2014, Chan et al., 2008, Chan and Vasconcelos, 2012, Cohen et al., 2000, Cui et al., 2007, Cui et al., 2008, Englebienne and Krose, 2010, Englebienne et al., 2009, Fanelli et al., 2011a, Fanelli et al., 2011b, Fod et al., 2002, Garc¿ıa et al., 2012, Gollan et al., 2011, Han et al., 2004, Harville, 2004, Heikkila and Silven, 1999, Hernández et al., 2011, Hernández-Sosa et al., 2011, Hou and Pang, 2011, Katabira et al., 2004, Kim et al., 2002, Lee et al., 2008, Leibe et al., 2008, Lorenzo-Navarro et al., 2013, Marcos et al., 2013, Mathews and Poigné, 2009, Moore et al., 2011, Nakamura et al., 2006, Oliver et al., 2012, Piccardi, 2004, Satta et al., 2013, Scheutz et al., 2004, Septian and Tao, 2006, Shotton et al., 2011, Spinello and Arras, 2011, Stauffer and Grimson, 1999, van Oosterhout et al., 2011, Velipasalar et al., 2006, Xia et al., 2011, Yahiaoui et al., 2010, Yu et al., 2007, Zeng and Ma, 2010, Zhan et al., 2008, Zhao et al., 2009 and Zivkovic and der Heijden, 2006.

Referencias

[Albiol et al., 2012]

Albiol, A., Albiol, A., Oliver, J., ., J.M., September 2012. Who is who at dif- ferent cameras: people re-identification using depth cameras. IET Computer Vision 6 (5), 378-387.

[Albiol et al., 2001]

Albiol, A., Mora, I., Naranjo, V., December 2001. Real-time high density peo- ple counter using morphological tools. IEEE Transactions on Intelligent Transportation Systems 2 (4), 204-218.

[Albiol and Silla, 2010]

Albiol, A., Silla, J., 2010. Statistical video analysis for crowds counting. En: Proceedings of the 16th IEEE international conference on Image Processing (ICIP). pp. 2569-2572.

[Andriluka et al., 2008]

Andriluka, M., Roth, S., Schiele, B., 2008. People-tracking-by-detection and people-detection-by-tracking. En: IEEE Conf. on Computer Vision and Pat- tern Recognition.

[Antic et al., 2010]

Antic, B., Letic, D., D. Culibrk, V.C., 2010. K-means based segmentation for real-time zenithal people counting. En: Proceedings of the 16th IEEE Inter- national Conference on the Image Processing (ICIP). pp. 2565-2568.

[Barandiaran et al., 2008]

Barandiaran, J., Murguia, B., Boto, F., 2008. Real-time people counting using multiple lines. En: Ninth International Workshop on Image Analysis for Multimedia Interactive Services. pp. 159-162.

[Barbosa et al., 2012]

Barbosa, B.I., Cristani, M., Bue, A.D., Bazzani, L., Murino, V., 2012. Re- identification with RGB-D sensors. En: 1st International Workshop on Re- Identification.

[Bellotto and Hu, 2009]

Bellotto, N., Hu, H., Feb- 2009. Multisensor-based human detection and trac- king for mobile service robots. IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics 39 (1), 167-181.

[Beymer, 2000]

Beymer, D., 2000. Person counting using stereo. En: Workshop on Human Mo- tion. pp. 127-133.

[Blanco et al., 2003]

Blanco, J., Burgard, W., Sanz, R., Fernandez, J., 2003. Fast face detection for mobile robots by integrating laser range data with vision. En: Proc. of the International Conference on Advanced Robotics (ICAR). pp. 953-958.

[Bozzoli et al., 2007]

Bozzoli, M., Cinque, L., Sangineto, E., 2007. A statistical method for people counting in crowded environments. En: 14th International Conference on Image Analysis and Processing.

[Brutzer et al., 2011]

Brutzer, S., Hoferlin, B., Heidemann, G., 2011. Evaluation of background sub- traction techniques for video surveillance. En: IEEE Conference on Compu- ter Vision and Pattern Recognition (CVPR). pp. 1937-1944.

[Camplani et al., 2014]

Camplani, M., del Blanco, C.R., Salgado, L., Jaureguizar, F., Garcí, N., Ja- nuary 2014. Advanced background modeling with RGB-D sensors through classifiers combination and inter-frame foreground prediction. Machine Vi- sion and Applications 25 (1), 122-136.

[Chan et al., 2008]

Chan, A.B., Liang, Z.-S. J., Vasconcelos, N., 2008. Privacy preserving crowd monitoring: Counting people without people models or tracking. En: Com- puter Vision and Pattern Recognition. pp. 1-7.

[Chan and Vasconcelos, 2012]

Chan, A.B., Vasconcelos, N., April 2012. Counting people with low-level fea- tures and bayesian regression. IEEE TRANSACTIONS ON IMAGE PRO- CESSING 21 (4), 2160-2177.

[Cohen et al., 2000]

Cohen, I., Garg, A., Huang, T., 2000. Vision-based overhead view person re- cognition. En: 15th International Conference on Pattern Recognition.

[Cui et al., 2007]

Cui, J., Zha, H., Zhao, H., Shibasaki, R., 2007. Laser-based detection and trac- king of multiple people in crowds. Computer Vision and Image Understan- ding 106, 300-312.

[Cui et al., 2008]

Cui, J., Zha, H., Zhao, H., Shibasaki, R., 2008. Multi-modal tracking of people using laser scanners and video camera. Image and Vision Computing 26 (2), 240-252.

[Englebienne and Krose, 2010]

Englebienne, G., Krose., B., 2010. Fast bayesian people detection. En: 22nd Benelux Conference on Artificial intelligence.

[Englebienne et al., 2009]

Englebienne, G., van Oosterhout, T., Krose, B., 2009. Tracking in sparse multi- camera setups using stereo vision. En: Third ACM/IEEE International Con- ference on Distributed Smart Cameras (ICDSC).

[Fanelli et al., 2011a]

Fanelli, G., Gall, J., Gool, L.V., 2011a. Real time head pose estimation with random regression forests. En: Computer Vision and Patter Recognition (CVPR).

[Fanelli et al., 2011b]

Fanelli, G., Weise, T., Gall, J., Gool, L.V., 2011b. Real time head pose esti- mation from consumer depth cameras. En: 33rd Annual Symposium of the German Association for Pattern Recognition (DAGM).

[Fod et al., 2002]

Fod, A., Howard, A., Mataric, M.J., May 2002. Laser-based people tracking. En: IEEE International Conference on Robotics and Automation (ICRA). Washington D.C., pp. 3024-3029.

[Garc¿ıa et al., 2012]

García, J., Gardel, A., Bravo, I., Lázaro, J.L., Martínez, M., Rodríguez, D., Oct-Diciembre 2012. Detección y seguimiento de personas basado en estereovisión y filtro de kalman. Revista Iberoamericana de Automática e Informática Industrial 9 (4).

[Gollan et al., 2011]

Gollan, B., Wally, B., Ferscha, A., 2011. Id management strategies for inter- active systems in multi-camera scenarios. En: 4th Conference on Context Awareness for Proactive Systems (CAPS). Budapest.

[Han et al., 2004]

Han, B., Comaniciu, D., Zhu, Y., Davis, L., 2004. Incremental density appro- ximation and kernel-based bayesian filtering for object tracking. En: IEEE Conf. on Computer Vision and Pattern Recognition. pp. 638-644.

[Harville, 2004]

Harville, M., 2004. Stereo person tracking with adaptive plan-view templates of height and occupancy statistics. Image and Vision Computing 22 (2), 127-142.

[Heikkila and Silven, 1999]

Heikkila, J., Silven, O., June 1999. A real-time system for monitoring of cyclists and pedestrians. En: IEEE Workshop on Visual Surveillance. Fort Collins, Colorado, pp. 82-90.

[Hernández et al., 2011]

Hernández, D., Castrillón, M., Lorenzo, J., 2011. People counting with re- identification using depth cameras. En: 4th International Conference on Ima- ging for Crime Detection and Prevention (ICDP).

[Hernández-Sosa et al., 2011]

Hernández-Sosa, D., Castrillón-Santana, M., Lorenzo-Navarro, J., 2011. Multi- sensor people counting. En: IbPRIA. pp. 321-328.

[Hou and Pang, 2011]

Hou, Y., Pang, G., 2011. People counting and human detection in a challen- ging situation. IEEE Transactions on Systems, Man and Cybernetics, Part A: Systems and Humans 41, 24-33.

[Katabira et al., 2004]

Katabira, K., Nakamura, K., Zhao, H., Shibasaki, R., November 22-26 2004. A method for counting pedestrians using a laser range scanner. En: 25th Asian Conference on Remote Sensing (ACRS 2004). Thailand.

[Kim et al., 2002]

Kim, J.W., Choi, K.S., Park, W.-S., Lee, J.-Y., Ko, S.J., September 2002. Robust real-time people tracking system for security. Intelligent Building Society (IBS) 2 (3), 184-190.

[Lee et al., 2008]

Lee, G.-G., ki Kim, H., Yoon, J.-Y., Kim, J.-J., Kim, W.-Y., 2008. Pedestrian counting using an IR line laser. En: International Conference on Convergen- ce and Hybrid Information Technology 2008.

[Leibe et al., 2008]

Leibe, B., Schindler, K., Cornelis, N., Gool:, L.J. V., 2008. Coupled object de- tection and tracking from static cameras and moving vehicles. IEEE Trans attern Anal. Mach. Intell. 30 (10), 1683-1698.

[Lorenzo-Navarro et al., 2013]

Lorenzo-Navarro, J., Castrillón-Santana, M., Hernández-Sosa, D., 2013. On the use of simple geometric descriptors provided by RGB-D sensors for re- identification. Sensors 13 (7), 8222-8238.

[Marcos et al., 2013]

Marcos, A., Pizarro, D., Marrón, M., Mazo, M., Abr 2013. Captura de mo- vimiento y reconocimiento de actividades para múltiples personas mediante un enfoque bayesiano. Revista Iberoamericana de Automática e Informática Industrial 10 (2).

[Mathews and Poigné, 2009]

Mathews, E., Poigné, A., 2009. Evaluation of a “art”destrian counting sys- tem based on echo state networks. EURASIP Journal on Embedded Systems 2009, 1-9 OI: http://dx.doi.org/10.1155/2009/352172.

[Moore et al., 2011]

Moore, B.E., Ali, S., Mehran, R., Shah, M., December 2011. Visual crowd surveillance through a hydrodynamics lens. Communications of the ACM 54 (12), 64-73.

[Nakamura et al., 2006]

Nakamura, K., Zhao, H., Shibasaki, R., K.S., Ohga, T., Suzukawa, N., 2006. Tracking pedestrians using multiple single-row laser range scanners and its reliability evaluation,. Systems and Computers in Japan 37, 1-11.

[Oliver et al., 2012]

Oliver, J., Albiol, A., Albiol, A., 2012. 3d descriptor for people re-identification n:, En: 21st International Conference on Pattern Recognition (ICPR).

[Piccardi, 2004]

Piccardi, M., 2004. Background subtraction techniques: a review. En: IEEE International Conference on Systems, Man and Cybernetics. pp. 3099-3104. Qiuyu, Z., Li;, T., Yiping, J., Wei-jun, D.,;1; 2010. A novel approach of counting people based on stereovision and dsp. En: The 2nd International Conference n Computer and Automation Engineering;1; (ICCAE).

[Satta et al., 2013]

Satta, R., Pala, F., Fumera, G., Roli, F., 2013. Real-time appearance-based person re-identification over multiple kinect cameras. En: 8th International Conference on Computer Vision Theory and Applications (VISAPP). Bar- celona, Spain.

[Scheutz et al., 2004]

Scheutz, M., McRaven, J., Cserey, G., 2004. Fast, reliable, adaptive, bimodal people tracking for indoor environments. En: Proceedings. 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2004. (IROS 2004). Vol. 2. pp. 1347-1352.

[Septian and Tao, 2006]

Septian, H., Tao, J., Tan, Y.-P., 5-8 Dec. 2006 2006. People counting by video segmentation and tracking. En: 9th International Conference on Control, Au- tomation, Robotics and Vision, 2006. ICARCV ‘06. Singapore, pp. 1-4.

[Shotton et al., 2011]

Shotton, J., Fitzgibbon, A., Cook, M., Sharp, T., Finocchio, M., Moore, R., Kipma, A., Blake, A., June 2011. Real-time human pose recognition in parts from a single depth image. En: Computer Vision and Pattern Recognition.

[Spinello and Arras, 2011]

Spinello, L., Arras, K.O., 2011. People detection in RGB-D data. En: Proc. of The International Conference on Intelligent Robots and Systems (IROS).

[Stauffer and Grimson, 1999]

Stauffer, Grimson, 1999. Adaptive background mixture models for real-time tracking. En: IEEE Computer Society Conference on Computer Vision and Pattern Recognition. pp. 246-252.

[van Oosterhout et al., 2011]

van Oosterhout, T., Bakkes, S., Kröse, B., 2011. Head detection in stereo da- ta for people counting and segmentation. En: International Conference on Computer Vision Theory and Applications (VISAPP). pp. 620-625.

[Velipasalar et al., 2006]

Velipasalar, S., li Tian, Y., Hampapur, A., July 2006. Automatic counting of interacting people by using a single uncalibrated camera. En: IEEE Interna- tional Conference on Multimedia and Expo. Toronto, ON, Canada.

[Xia et al., 2011]

Xia, L., Chen, C.-C., Aggarwal, J.K., June 2011. Human detection using depth information by kinect. En: International Workshop on Human Activity Un- derstanding from 3D Data in conjunction with CVPR (HAU3D). Colorado Springs, CO.

[Yahiaoui et al., 2010]

Yahiaoui, T., Khoudour, L., Meurie, C., July 2010. Real-time passenger coun- ting in buses using dense stereovision. J. Electron. Imaging 20.

[Yu et al., 2007]

Yu, H., Liu, J., Liu, J., 2007. 3d feature extraction of head based on target region matching. En: Proceedings of the International Conference on Computatio- nal Intelligence and Security. pp. 366-370.

[Zeng and Ma, 2010]

Zeng, C., Ma, H., 2010. Robust head-shoulder detection by pca-based multile- vel HOG-LBP detector for people counting. En: 20th International Confe- rence on Pattern Recognition (ICPR). Istambul, pp. 2069-2072.

[Zhan et al., 2008]

Zhan, B., Monekosso, D.N., Remagnino, P., Velastin, S.A., Xu, L.-Q., 2008. Crowd analysis: a survey. Machine Vision and Applications 19, 345-357.

[Zhao et al., 2009]

Zhao, X., Dellandréa, E., Chen, L., 2009. A people counting system based on face detection and tracking in a video. En: Advanced Video and Signal Ba- sed Surveillance.

[Zivkovic and der Heijden, 2006]

Zivkovic, Z., der Heijden, F., 2006. Effcient adaptive density estimation per image pixel for the task of background subtraction. Pattern Recognition Let- ters 27, 773-780.

Indexada en:

Síguenos:

Indexada en:

Síguenos:

Suscríbase a la newsletter