Abstract
Closed circuit television cameras (CCTV) are widely used in monitoring. This paper presents an intelligent CCTV crowd counting system based on two algorithms that estimate the density of each pixel in each frame and use it as a basis for counting people. One algorithm uses scale-invariant feature transform (SIFT) features and clustering to represent pixels of frames (SIFT algorithm) and the other uses features from accelerated segment test (FAST) corner points with SIFT features (SIFT-FAST algorithm). Each algorithm is designed using a novel combination of pixel-wise, motion-region, grid map, background segmentation using Gaussian mixture model (GMM) and edge detection. A fusion technique is proposed and used to validate the accuracy by combining the result of the algorithms at frame level. The proposed system is more practical than the state of the art regression methods because it is trained with a small number of frames so it is relatively easy to deploy. In addition, it reduces the training error, set-up time, cost and open the door to develop more accurate people detection methods. The University of California (UCSD) and Mall datasets have been used to test the proposed algorithms. The mean deviation error, mean squared error and the mean absolute error of the proposed system are less than 0.1, 16.5 and 3.1, respectively, for the Mall dataset and less than 0.07, 5.5 and 1.9, respectively, for UCSD dataset.
Similar content being viewed by others
References
Adegboye AO (2013) Single pixel robust approach for background subtraction for fast people-counting and direction estimation. University of Pretoria, Dissertation
Adegboye A, Hancke G, Jr GH (2012) Single-pixel approach for fast people counting and direction estimation. South. Africa Telecommun, Networks Appl
Al-Zaydi ZQH, Ndzi DL, Yang Y, Kamarudin ML (2016) An adaptive people counting system with dynamic features selection and occlusion handling. J Vis Commun Image Represent 39:218–225. doi:10.1016/j.jvcir.2016.05.018
Antonini G, Thiran JP (2004) Trajectories clustering in ICA space an application to automatic counting of pedestrians in video sequences. Adv. Concepts Intell. Vis, Syst
Benezeth Y, Jodoin P-M, Emile B et al (2010) Comparative study of background subtraction algorithms. J Electron Imaging 19:33003. doi:10.1117/1.3456695
Berndt D, Clifford J (1994) Using dynamic time warping to find patterns in time series. Report, AAAI
Bottesch T, Markus K, Kaechele M, Ulm U (2016) Speeding up k -means by approximating Euclidean distances via block vectors. Int. Conf. Mach. Learn, In, pp. 2578–2586
Bouwmans T, El Baf F, Vachon B (2008) Background modeling using mixture of Gaussians for foreground detection - a survey. Recent Patents Comput Sci 1:219–237. doi:10.2174/2213275910801030219
Brostow GJ, Cipolla R (2006) Unsupervised Bayesian detection of independent motion in crowds. IEEE Conf Comput Vis Pattern Recognit. doi:10.1109/CVPR.2006.320
Çelik H, Hanjalić A, Hendriks EA (2006) Towards a robust solution to people counting. Int. Conf. Image Process, In, pp. 2401–2404
Chan AB, Vasconcelos N (2009) Bayesian poisson regression for crowd counting. In: IEEE Int. Conf, Comput. Vis. IEEE, pp. 545–551
Chan AB, Vasconcelos N (2012) Counting people with low-level features and bayesian regression. IEEE Trans Image Process 21:2160–2177. doi:10.1109/TIP.2011.2172800
Chan AB, Liang ZSJ, Vasconcelos N (2008) Privacy preserving crowd monitoring: counting people without people models or tracking. IEEE Conf Comput Vis Pattern Recognit. doi:10.1109/CVPR.2008.4587569
Chan A, Morrow M, Vasconcelos N (2009) Analysis of crowded scenes using holistic properties. In: Perform. Eval, Track. Surveill. Work. IEEE, pp. 101–108
Chen K, Kamarainen J-K (2014) Learning to count with back-propagated information. Int. Conf. Pattern Recognit. IEEE, In, pp. 4672–4677
Chen K, Loy CC, Gong S, Xiang T (2012) Feature Mining for Localised Crowd Counting. Br Mach Vis Conf. doi:10.5244/C.26.21
Chen K, Gong S, Xiang T, Loy CC (2013) Cumulative attribute space for age and crowd density estimation. In: IEEE Conf. Comput, Vis. Pattern Recognit, pp. 2467–2474
Cheriyadat AM, Bhaduri BL, Radke RJ (2008) Detecting multiple moving objects in crowded environments with coherent motion regions. IEEE Conf Comput Vis Pattern Recognit Work. doi:10.1109/CVPRW.2008.4562983
Cho SY, Chow TW (1999) A fast neural learning vision system for crowd estimation at underground stations platform. Neural Process Lett 10(2):111–120
Cho S-Y, Chow T, Leung C (1999) A neural-based crowd estimation by hybrid global learning algorithm. IEEE Trans Syst Man, Cybern Part B Cybern 29:535–541. doi:10.1109/3477.775269
Chow TWS, Yam JYF, Cho SY (1999) Fast training algorithm for feedforward neural networks: application to crowd estimation at underground stations. Artif Intell Eng 13:301–307. doi:10.1016/S0954-1810(99)00016-3
Conte D, Foggia P, Percannella G et al (1743–1746) (2010) counting moving people in videos by salient points detection. Int. Conf. Pattern Recognit. pp, In
Conte D, Foggia P, Percannella G et al (2010) A method for counting people in crowded scenes. In: IEEE Int. Conf, Adv. Video Signal Based Surveill, pp. 225–232
Conte D, Foggia P, Percannella G, Vento M (2013) Counting moving persons in crowded scenes. Mach Vis Appl 24:1029–1042. doi:10.1007/s00138-013-0491-3
Dalal N, Triggs B (2005) Histograms of oriented gradients for human detection. In: IEEE Conf. Comput, Vis. Pattern Recognit, pp. 886–893
Davies A, Yin JH, Velastin S (1995) Crowd monitoring using image processing. Electron Commun Eng J. doi:10.1049/ecej:19950106
Dollar P, Belongie S, Perona P (2010) The fastest pedestrian detector in the west. Br Mach Vis Conf. doi:10.5244/C.24.68
Felzenszwalb PF, Girshick RB, McAllester D, Ramanan D (2010) Object detection with discriminative trained part based models. IEEE Trans Pattern Anal Mach Intell 32:1627–1645
Fradi H, Dugelay JL (2012) Low level crowd analysis using frame-wise normalized feature for people counting. Int. Work. Inf. Forensics Secur, In, pp. 246–251
Gao L, Wang Y, Ye X, Wang J (2016) Crowd Pedestrian Counting Considering Network Flow Constraints in Videos. arXiv Prepr
Ge W, Collins RT (2009) Marked point processes for crowd counting. In: Comput. Vis, Pattern Recognit. Work. IEEE, pp. 2913–2920
Hafeezallah A, Abu-Bakar S (2016) Crowd counting using statistical features based on curvelet frame change detection. Multimed Tools Appl. doi:10.1007/s11042-016-3869-1
Harville M (2002) Stereo person tracking with adaptive plan-view statistical templates. Proc. ECCV Work. Stat. Methods Video Process, In, pp. 67–72
Hashimoto K, Morinaka K, Yoshiike N et al (1997) People count system using multi-sensing application. Int. Solid State Sensors Actuators Conf, In, pp. 1291–1294
Hou YL, Pang GKH (2011) People counting and human detection in a challenging situation. IEEE Trans Syst Man, Cybern Part ASystems Humans 41:24–33. doi:10.1109/TSMCA.2010.2064299
Hu X, Zheng H, Chen Y, Chen L (2015) Dense crowd counting based on perspective weight model using a fisheye camera. Int J Light Electron Opt 126:123–130. doi:10.1016/j.ijleo.2014.08.132
Hu Y, Chang H, Nian F et al (2016) Dense crowd counting from still images with convolutional neural networks. J Vis Commun Image Represent 38:530–539. doi:10.1016/j.jvcir.2016.03.021
Huang X, Zou Y, Wang Y (2016) Cost-sensitive sparse linear regression for crowd counting with imbalanced training data. IEEE Int. Conf. Multimed, Expo
Intelcom DILAX (2015) Public Transport https://www.dilax.com/. Accessed 1 Oct 2016
Jeong CY, Choi S, Han SW (2013) A method for counting moving and stationary people by interest point classification. In: IEEE Int. Conf, Image Process. IEEE, pp. 4545–4548
Joshi NS, Choubey NS (2014) Comparison of traditional approach for edge detection with soft computing approach. Int J Comput Appl 96:17–23
Kaur G, Virk IS (2014) Edge detection through fuzzy system using type I format. Int J Comput Appl 102:24–27
Kilambi P, Ribnick E, Joshi AJ et al (2008) Estimating pedestrian counts in groups. Comput Vis Image Underst 110:43–59. doi:10.1016/j.cviu.2007.02.003
Kong D, Gray D, Tao H (2005) Counting pedestrians in crowds using viewpoint invariant training. Procedings Br Mach Vis Conf. doi:10.5244/C.19.63
Kong D, Gray D, Tao H (2006) A viewpoint invariant approach for crowd counting. Int. Conf. Pattern Recognit, In, pp. 1187–1190
Leibe B, Seemann E, Schiele B (2005) Pedestrian detection in crowded scenes. In: IEEE Conf. Comput, Vis. Pattern Recognit, pp. 878–885
Lempitsky V, Zisserman A (2010) Learning to count objects in images. Adv. Neural Inf. Process. Syst, In, pp. 1324–1332
Li J, Huang L, Liu C (2011) Robust people counting in video surveillance: dataset and system. Int. Conf. Adv. Video Signal Based Surveill, In, pp. 54–59
Lin S, Chen J, Chao H (2001) Estimation of number of people in crowded scenes using perspective transformation. IEEE Trans Syst Man Cybern 31:645–654
Loy C, Chen K, Gong S, Xiang T (2013) Crowd counting and profiling: methodology and evaluation. Model. Simul. Vis. Anal. Crowds. Springer New York, In, pp. 347–382
Ltd B (2013) Use CCTV to Count People http://www.videoturnstile.com/. Accessed 1 Oct 2016
Ma R, Li L, Huang W, Tian Q (2004) On pixel count based crowd density estimation for visual surveillance. In: IEEE Conf. Cybern, Intell. Syst. IEEE, pp. 1–3
Ma H, Zeng C, Ling CX (2012) A reliable people counting system via multiple cameras. ACM Trans Intell Syst Technol 3:1–22. doi:10.1145/2089094.2089107
Merad D, Aziz KE, Thome N (2010) Fast people counting using head detection from skeleton graph. Adv. Video Signal Based Surveill. IEEE, In, pp. 233–240
Norris C, Mccahill M, Wood D (2004) Editorial. The growth of CCTV: a global perspective on the international diffusion of video surveillance in publicly accessible space. Surveill Soc 2:110–135
Rabaud V, Belongie S (2006) Counting crowded moving objects. In: IEEE Conf. Comput, Vis. Pattern Recognit, pp. 705–711
Rao AS, Gubbi J, Marusic S, Palaniswami M (2015) Estimation of crowd density by clustering motion cues. Vis Comput 31:1533–1552. doi:10.1007/s00371-014-1032-4
Rodriguez M, Superieure EN, Laptev I et al (2011) Density-aware person detection and tracking in crowds. Int. Conf. Comput. Vis. IEEE, In, pp. 2423–2430
Ryan DA (2013) Crowd monitoring using computer vision. Queensland University of Technology, Dissertation
Ryan D, Denman S, Fookes C, Sridharan S (2009) Crowd counting using multiple local features. Digit. Image Comput. Tech. Appl. IEEE, In, pp. 81–88
Ryan D, Denman S, Fookes C, Sridharan S (2014) Scene invariant multi camera crowd counting. Pattern Recogn Lett 44:98–112. doi:10.1016/j.patrec.2013.10.002
Ryan D, Denman S, Sridharan S, Fookes C (2015) An evaluation of crowd counting methods, features and regression models. Comput Vis Image Underst 130:1–17. doi:10.1016/j.cviu.2014.07.008
Saleh SAM, Suandi SA, Ibrahim H (2015) Recent survey on crowd density estimation and counting for visual surveillance. Eng Appl Artif Intell 41:103–114. doi:10.1016/j.engappai.2015.01.007
Shbib R, Zhou S, Ndzi D, Al-kadhimi K (2013) Distributed monitoring system based on weighted data fusing model. Am J Soc Issues Humanit 3:53–62
ShopperTrak (2013) ShopperTrak Solutions http://www.shoppertrak.com/. Accessed 1 Oct 2016
Shrivakshan GT, Chandrasekar C (2012) A Comparison of various Edge Detection Techniques used in Image Processing Int J Comput Sci Issues:9
Sidla O, Lypetskyy Y, Brändle N, Seer S (2006) Pedestrian detection and tracking for counting applications in crowded situations. IEEE Int Conf Video Signal Based Surveill. doi:10.1109/AVSS.2006.91
Sobral A, Vacavant A (2014) A comprehensive review of background subtraction algorithms evaluated with synthetic and real videos. Comput Vis Image Underst 122:4–21. doi:10.1016/j.cviu.2013.12.005
Tang NC, Lin Y-Y, Weng M, Liao HM (2015) Cross-camera knowledge transfer for Multiview people counting. IEEE Trans Image Process 24:80–93. doi:10.1109/TIP.2014.2363445
Technology A (2013) Our customers http://www.peoplecounting.co.uk/our-customers. Accessed 1 Oct 2016
Topkaya IS, Erdogan H, Porikli F (2014) Counting people by clustering person detector outputs. In: IEEE Int. Conf, Adv. Video Signal Based Surveill. IEEE, pp. 313–318
Tu J, Zhang C, Hao P (2013) Robust real-time attention-based head-shoulder detection for video surveillance. In: IEEE Int. Conf, Image Process. IEEE, pp. 3340–3344
Tuzel O, Porikli F, Meer P (2008) Pedestrian detection via classification on Riemannian manifolds. IEEE Trans Pattern Anal Mach Intell. doi:10.1109/TPAMI.2008.75
Wang M (2014) Data assimilation for agent-based simulation of smart environment. Georgia State University, Dissertation
Wang M, Wang X (2011) Automatic adaptation of a generic pedestrian detector to a specific traffic scene. In: IEEE Conf. Comput, Vis. Pattern Recognit, pp. 3401–3408
Wang J, Fu W, Liu J et al (2014) Spatiotemporal group context for pedestrian counting. IEEE Trans Circuits Syst Video Technol 24:1620–1630
Wu B, Nevatia R (2007) Detection and tracking of multiple, partially occluded humans by Bayesian combination of edgelet based part detectors. Int J Comput Vis 75:247–266. doi:10.1007/s11263-006-0027-7
Xiaohua L, Lansun S, Huanqin L (2006) Estimation of crowd density based on wavelet and support vector machine. Trans Inst Meas Control 28:299–308. doi:10.1191/0142331206tim178oa
Xing X, Wang K, Lv Z (2015) Fusion of gait and facial features using coupled projections for people identification at a distance. Signal Process Lett 22:2349–2353
Xu B, Qiu G (2016) Crowd density estimation based on rich features and random projection Forest. IEEE Winter Appl. Comput. Vis, In, pp. 1–8
Zhang J, Tan B, Sha F, He L (2011) Predicting pedestrian counts in crowded scenes with rich and high-dimensional features. IEEE Trans Intell Transp Syst 12:1037–1046. doi:10.1109/TITS.2011.2132759
Zhang C, Li H, Wang X (2015a) Cross-scene crowd counting via deep convolutional neural networks. Proc IEEE Conf Comput Vis Pattern Recognit. doi:10.1109/CVPR.2015.7298684
Zhang Z, Wang M, Geng X (2015b) Crowd counting in public video surveillance by label distribution learning. Neurocomputing 166:151–163. doi:10.1016/j.neucom.2015.03.083
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Al-Zaydi, Z.Q.H., Ndzi, D.L., Kamarudin, M.L. et al. A robust multimedia surveillance system for people counting. Multimed Tools Appl 76, 23777–23804 (2017). https://doi.org/10.1007/s11042-016-4156-x
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-016-4156-x