Real-time crowd counting method from video stream based on GPU

doi:10.11772/j.issn.1001-9081.2017.01.0145

Journal of Computer Applications ›› 2017, Vol. 37 ›› Issue (1): 145-152.DOI: 10.11772/j.issn.1001-9081.2017.01.0145

Previous Articles Next Articles

Real-time crowd counting method from video stream based on GPU

JI Lina¹, CHEN Qingkui^1,2, CHEN Yuanjing¹, ZHAO Deyu², FANG Yuling², ZHAO Yongtao¹

1. College of Optical-Electrical and Computer Engineering, University of Shanghai Science and Technology, Shanghai 200093, China;
2. College of Business, University of Shanghai Science and Technology, Shanghai 200093, China

Received:2016-07-30 Revised:2016-08-08 Online:2017-01-10 Published:2017-01-09
Supported by:
This work is partially supported by the National Natural Science Foundation of China (61572325, 60970012), the Ministry of Education Doctoral Funds for Ph.D Supervisor of China (20113120110008), Shanghai Key Science and Technology Project (14511107902), Shanghai Engineering Research Center Project (GCZX14014), Shanghai Leading Academic Discipline Project (XTKX2012), Shanghai Special Fund for Research Base (C14001).

基于GPU的视频流人群实时计数

姬丽娜¹, 陈庆奎^1,2, 陈圆金¹, 赵德玉², 方玉玲², 赵永涛¹

1. 上海理工大学光电信息与计算机工程学院, 上海 200093;
2. 上海理工大学管理学院, 上海 200093

通讯作者: 姬丽娜
作者简介:姬丽娜(1990-),女,河南信阳人,硕士研究生,主要研究方向:计算机图形图像处理、并行计算、模式识别;陈庆奎(1966-),男,上海人,教授,博士生导师,博士,CCF会员,主要研究方向:网络计算、并行计算、物联网;陈圆金(1991-),男,河南信阳人,硕士研究生,主要研究方向:单光子探测器;赵德玉(1978-),男,山东临沂人,博士研究生,主要研究方向:高性能计算、GPU架构、CPU-GPU异构集群调度;方玉玲(1990-),女,河南信阳人,博士研究生,主要研究方向:高性能计算、GPU架构、CPU-GPU异构集群调度;赵永涛(1991-),男,河北邯郸人,硕士研究生,主要研究方向:模式识别、并行计算。
基金资助:
国家自然科学基金资助项目（61572325，60970012）；高等学校博士学科点专项科研博导基金资助项目（20113120110008）；上海重点科技攻关项目（14511107902）；上海市工程中心建设项目（GCZX14014）；上海市一流学科建设项目（XTKX2012）；沪江基金研究基地专项（C14001）。

Abstract

Abstract: Focusing on low counting accuracy caused by serious occlusions and abrupt illumination variations, a new real-time statistical method based on Gaussian Mixture Model (GMM) and Scale-Invariant Feature Transform (SIFT) features for video crowd counting was proposed. Firstly, the moving crowd were detected by using GMM-based motion segment method, and then the Gray Level Co Occurrence Matrix (GLCM) and morphological operations were applied to remove small moving objects of background and the dense noise in non-crowd foreground. Considering the high time-complexity of GMM algorithm, a novel parallel model with higher efficiency was proposed. Secondly, the SIFT feature points were acted as the basis of crowd statistics, and the execution time was reduced by using feature exaction based on binary image. Finally, a novel statistical analysis method based on crowd features and crowd number was proposed. The data sets with different level of crowd number were chosen to train and get the average feature number of a single person, and the pedestrians with different densities were counted in the experiment. The algorithm was accelerated by using multi-stream processors on Graphics Processing Unit (GPU) and the analysis about efficiently scheduling the tasks on Compute Unified Device Architecture (CUDA) streams in practical applications was conducted. The experimental results indicate that the speed is increased by 31.5% compared with single stream, by 71.8% compared with CPU.

Key words: video surveillance, Graphics Processing Unit (GPU) parallel computing, crowd counting, Scale-Invariant Feature Transform (SIFT), Gaussian Mixture Model (GMM), Compute Unified Device Architecture (CUDA)

摘要： 为了解决人群遮挡严重、光照突变等恶劣环境下人群计数准确率低的问题，提出基于混合高斯模型（GMM）和尺度不变特征变换（SIFT）特征的人群数量统计分析新方法。首先，基于GMM提取运动人群，并采用灰度共生矩阵（GLCM）和形态学方法去除背景中移动的小物体和较密集的噪声等非人群前景，针对GMM算法提出了一种效率较高的并行模型；接着，检测运动人群的SIFT特征点作为人群统计的基础，基于二值图像的特征提取大大减少了执行时间；最后，提出基于人群特征数和人群数量进行统计分析的新方法，选择不同等级的人群数量的数据集分别进行训练，统计得出平均单个特征点数，并对不同密度的行人进行计数实验。算法采用基于GPU多流处理器进行加速，并针对所提算法在统一计算设备架构（CUDA）流上任务的有效调度的方法进行分析。实验结果显示，相比单流提速31.5%，相比CPU提速71.8%。

关键词: 视频监控, GPU并行计算, 人群计数, 尺度不变特征变换, 混合高斯模型, 统一计算设备架构

CLC Number:

TP391.413

JI Lina, CHEN Qingkui, CHEN Yuanjing, ZHAO Deyu, FANG Yuling, ZHAO Yongtao. Real-time crowd counting method from video stream based on GPU[J]. Journal of Computer Applications, 2017, 37(1): 145-152.

姬丽娜, 陈庆奎, 陈圆金, 赵德玉, 方玉玲, 赵永涛. 基于GPU的视频流人群实时计数[J]. 计算机应用, 2017, 37(1): 145-152.

References

[1] ZENG C B, MA H D. Robust head-shoulder detection by PCA-based multilevel HOG-LBP detector for people counting[C]//ICPR 2010:Proceedings of the 201020th International Conference on Pattern Recognition. Piscataway, NJ:IEEE, 2010:2069-2072.
[2] DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[C]//CVPR' 05:Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2005:886-893.
[3] LI M, ZHANG Z X, HUANG K Q, et al. Estimating the number of people in crowded scenes by MID based foreground segmentation and head-shoulder detection[C]//ICPR 2008:Proceedings of the 19th International Conference on Pattern Recognition. Piscataway, NJ:IEEE, 2008:1-4.
[4] ALBIOL A, SILLA M J, ALBIOL A, et al. Video analysis using corner motion statistics[C]//PETS 2009:Proceedings of the 2009 IEEE International Workshop on Performance Evaluation of Tracking and Surveillance. Washington, DC:IEEE Computer Society, 2009:31-38.
[5] CHAN A B, VASCONCELOS N. Counting people with low-level features and Bayesian regression[J]. IEEE Transactions on Image Processing, 2012, 21(4):2160-2177.
[6] KONG D, GRAY D, TAO H. A viewpoint invariant approach for crowd counting[C]//ICPR 2006:Proceedings of the 18th International Conference on Pattern Recognition. Piscataway, NJ:IEEE, 2006:1187-1190.
[7] CHI Y J, SUGIL C, SEUNG W H. A method for counting moving and stationary people by interest point classification[C]//ICIP 2013:Proceedings of the 2013 IEEE International Conference on Image Processing. Piscataway, NJ:IEEE, 2013:4545-4548.
[8] CONTE D, FOGGIA P, PERCANNELLA G, et al. Counting moving people in videos by salient points detection[C]//EUSIPCO 2012:Proceedings of the 20th International Conference on Pattern Recognition. Washington, DC:IEEE Computer Society, 2010:1743-1746.
[9] FRADI H, DUGELAY J L. People counting system in crowded scenes based on feature regression[C]//ICPR 2010:Proceedings of the 20th European Signal Processing Conference. Bucharest, Romania:[s.n.], 2012:136-140.
[10] CONTE D, FOGGIA P, PERCANNELLA G, et al. A method for counting people in crowded scenes[C]//AVSS 2010:Proceedings of the 2010 IEEE International Conference on Advanced Video and Signal Based Surveillance. Piscataway, NJ:IEEE, 2010:225-232.
[11] HOU Y L, PANG G K H. People counting and human detection in a challenging situation[J]. IEEE Transactions on Systems, Man, and Cybernetics, Part A:Systems and Humans, 2011, 41(1):24-33.
[12] HOU Y L, PANG G K H. Automated people counting at a mass site[C]//Proceedings of the 2008 IEEE International Conference on Automation and Logistics. Piscataway, NJ:IEEE, 2008:464-469.
[13] LI H, ACHIM A, BULL D R. GMM-based efficient foreground detection with adaptive region update[C]//ICIP 2009:Proceedings of the 16th IEEE International Conference on Image Processing. Piscataway, NJ:IEEE, 2009:3145-3148.
[14] 姬丽娜,陈庆奎,赵永涛,等.一种基于GLCM的运动目标检测新方法[J].太原理工大学学报,2015,46(6):719-726.(JI L N, CHEN Q K, ZHAO Y T, et al. A novel method of moving object detection based on GLCM[J]. Journal of Taiyuan University of Technology, 2015, 46(6):719-726.)
[15] SANDERS J, KANDROT E. GPU高性能编程CUDA实战[M].聂雪军,译.北京:机械工业出版社,2011:139-151.(SANDERS J, KANDROT E. CUDA by Example:An Introduction to General-Purpose GPU Programming[M]. NIE X J, translated. Beijing:China Machine Press, 2011:139-151.)
[16] HARRIS C, STEPHENS M. A combined corner and edge detector[C]//Proceedings of the 4th Alvey Vision Conference. Manchester:Alvey Vision Club, 1988:147-151.
[17] 梁荣华,刘向东,马祥音,等.基于SURF的高密度人群计数方法[J].计算机辅助设计与图形学学报,2012,24(12):1568-1575.(LIANG R H, LIU X D, MA X Y, et al. High-density crowd counting method based on SURF feature[J]. Journal of Computer-Aided Design & Computer Graphics, 2012, 24(12):1568-1575.)

Real-time crowd counting method from video stream based on GPU

基于GPU的视频流人群实时计数

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics

[1]	XIE Wenbo, WEI Yongzhuang, LIU Zhenghong. Parallel implementation and analysis of SKINNY encryption algorithm using CUDA [J]. Journal of Computer Applications, 2021, 41(4): 1136-1141.
[2]	YANG Xianfeng, GUI Hongjun, FU Chunchang. F-X domain predictive filtering parallel algorithm based on compute unified device architecture [J]. Journal of Computer Applications, 2021, 41(2): 486-491.
[3]	FU Qianhui, LI Qingkui, FU Jingnan, WANG Yu. Dense crowd counting model based on spatial dimensional recurrent perception network [J]. Journal of Computer Applications, 2021, 41(2): 544-549.
[4]	DU Peide, YAN Hua. Crowd counting network based on multi-scale spatial attention feature fusion [J]. Journal of Computer Applications, 2021, 41(2): 537-543.
[5]	LI Ziqiang, WANG Zhengyong, CHEN Honggang, LI Linyi, HE Xiaohai. Video abnormal behavior detection based on dual prediction model of appearance and motion features [J]. Journal of Computer Applications, 2021, 41(10): 2997-3003.
[6]	YANG Huihua, ZHANG Tianyu, LI Lingqiao, PAN Xipeng. Target recognition algorithm for urban management cases by mobile devices based on MobileNet [J]. Journal of Computer Applications, 2019, 39(8): 2475-2479.
[7]	LU Jingang, ZHANG Li. Crowd counting model based on multi-scale multi-column convolutional neural network [J]. Journal of Computer Applications, 2019, 39(12): 3445-3449.
[8]	LI Jialiang, JIANG Pinqun. Image stitching by combining deformation function and power function weight [J]. Journal of Computer Applications, 2019, 39(10): 3060-3064.
[9]	YANG Chunde, CHENG Yanfei. Fast image mosaic algorithm based on adaptive elimination of stitching seam and panorama alignment [J]. Journal of Computer Applications, 2019, 39(10): 3053-3059.
[10]	CAO Jinmeng, NI Rongrong, YANG Biao. Crowd counting using multi-scale multi-task convolutional neural network [J]. Journal of Computer Applications, 2019, 39(1): 199-204.
[11]	LIN Lang, WANG Rangding, YAN Diqun, LI Can. Playback speech detection algorithm based on modified cepstrum feature [J]. Journal of Computer Applications, 2018, 38(6): 1648-1652.
[12]	WU Liang, HE Yi, MEI Xue, LIU Huan. Online behavior recognition using space-time interest points and probabilistic latent-dynamic conditional random field model [J]. Journal of Computer Applications, 2018, 38(6): 1760-1764.
[13]	WEI Yuxi, WU Yueqing, TAO Pan, YAO Yu. Left ventricle segmentation in transesophageal echocardiography based on supervised descent method [J]. Journal of Computer Applications, 2018, 38(2): 545-549.
[14]	YU Xinrong, LI Zhihua, YAN Chengyu, LI Shuangli. High efficient virtual machines consolidation method in cloud data center [J]. Journal of Computer Applications, 2018, 38(2): 550-556.
[15]	TAO Zhiyong, LIU Xiaofang, WANG Hezhang. Clustering algorithm of Gaussian mixture model based on density peaks [J]. Journal of Computer Applications, 2018, 38(12): 3433-3437.