[1]吴冬梅,王静,李白萍,等.基于改进SURF算法的大规模群体人数统计[J].西安科技大学学报,2015,(05):650-655.[doi:10.13800/j.cnki.xakjdxxb.2015.0520]
 WU Dong-Mei,WANG Jing,LI Bai-Ping,et al.Large crowd count based on improved SURF algorithm[J].Journal of Xi'an University of Science and Technology,2015,(05):650-655.[doi:10.13800/j.cnki.xakjdxxb.2015.0520]
点击复制

基于改进SURF算法的大规模群体人数统计(/HTML)
分享到:

西安科技大学学报[ISSN:1672-9315/CN:61-1434/N]

卷:
期数:
2015年05期
页码:
650-655
栏目:
出版日期:
2015-10-08

文章信息/Info

Title:
Large crowd count based on improved SURF algorithm
作者:
吴冬梅王静李白萍郭婷
西安科技大学 通信与信息工程学院,陕西 西安 710054
Author(s):
WU Dong-MeiWANG JingLI Bai-PingGUO Ting
College of Communication and Information Engineering,Xi'an University of Science and Technology,Xi'an 710054,China
关键词:
人数统计 SURF 灰度共生矩阵 透视矫正 支持向量回归
Keywords:
crowd count SURF gray level co-occurrence matrix perspective-correct support vector regression
分类号:
TN 911.73
DOI:
10.13800/j.cnki.xakjdxxb.2015.0520
文献标志码:
A
摘要:
为了在高密度大规模群体人数统计的问题上有效的克服遮挡与摄像机透视畸形带来的影响,文中采用了一种基于线性内插透视矫正的SURF(Speeded Up Robust Feature)算法。首先,采用背景差与滑动平均相结合的方式得到人群前景,并通过对二值前景图像的形态学处理进行去噪。其次,对获取到的前景图像进行多特征提取,将传统的灰度共生矩阵特征与SURF算法特征相结合,并通过线性内插权值的透视矫正方法进行摄像畸形矫正,将矫正后的特征值组成了表征人群数目特征的特征向量。从而减少了深度信息丢失而引起的误差,得到了优化的人群特征向量; 最后,通过支持向量回归的方式拟合出人群人数统计模板,以此预测监控区域的人数。实验表明文中方法具有较高的准确性,较传统SURF算法准确率有了很高的提升。
Abstract:
The SURF based on the method of Linear Interpolation for camera distortion calibration is adopted for high-density crowd counting.The eigenvalues are built on the Gray Level Co-occurrence Matrix(GLCM)features and the SURF features.To get the foreground image,firstly,gray and smooth the input image.Then getting foreground image by background subtraction operation and moving average method.And also morphology processing was performed on the binary image to eliminate noise.And then,extracting feature parameters of foreground image.Though the method of linear interpolation,weight values are interpolated to reducethe error,which is caused bycamera distortion calibration.Linear interpolation weights perspective correction method is considered for camera deformity correction.The optimized crowd feature vector can be obtained then.Through the method of support vector regression(SVR),the crowd number can be forecasted by the training model.The experiment result shows that the method of this paper has a higher accuracy than the previous methods.

参考文献/References:

[1] 刘向东.高密度人群计数方法的研究与应用[D].杭州:浙江工业大学,2012. LIU Xiang-dong.Research and application of high-density counting method[D].Hangzhou:Zhejiang University of Technology,2012.
[2] 杨 华,苏 航,郑世宝.大规模群体密度估计算法[J].视频应用与工程,2010,34(5):113-116. YANG Hua,SU Hang,ZHENG Shi-bao.Large-scale crowd density estimation[J].Video Application & Project,2010,34(5):113-116.
[3] 吕济民,曾昭贤,张茂军.基于非最大抑制聚类的视频人数估测方法[J].模式识别与人工智能,2012,25(1):150-156. LV Ji-min,ZENG Zhao-xian,ZHANG Mao-jun.Counting pedestrains in video sequences based on non-maxima suppression clustering[J].Pattern Relognition and Artificial Infeuigence,2012,25(1):150-156.
[4] Cho S Y,Chow T W S,Leung C T.A neural-based crowd estimation by hybrid global learning algorithm[J].IEEE.Transactions on Systems,Man,and Cybernetics-Part B:Cybernetics,1999,29(4):535-541.
[5] Nevatia R,Zhao T.Tracking multiple humans in crowded environments[C]//IEEE International Conference on Computer Vision and Pattern Recognition,Washington DC,USA,2004
[6] Damian Roqueiro A,Valery A,Petrushin.Counting people using video cameras[J].International Journal of Parallel.Emergent and Distributed Systems,2007:193-209
[7] Proceedings of eleventh IEEE international workshop on performance evaluation of tracking and surveillance(PETS 2009)[R].IEEE Computer Science,2009.
[8] Conte D,Foggia P,Percannella G,et al.A method for counting moving people in video surveillance videos[J].EURASIP Journal on Advances in Signal Processing,2010,5(1):1-8.
[9] HOU Ya-li,Pang 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):245-253.
[10]Bay H,Ess A,Tuytelaars T,et al.SURF:speeded up robust features[J].Computer Vision and Image Understanding,2008,110(3):346-359.
[11]杨德坤,侯德文,步亚东.灰度共生矩阵在纹理特征提取中的发展[J].信息系统工程,2012,15(1):22-25. YANG De-kun,HOU De-wen,BU Ya-dong.Development of gray co-occurrence matrix in texture feature extraction[J].Information system engineering,2012,15(1):22-25.
[12]Rahmalan H,Nixon M,Carter J.On crowd density estimation for surveillance[J].Institution of Engineering and Technology,2006,8(12):315-325.
[13]王雅琳.基于灰度共生矩阵的人群密度估计算法研究[D].西安:西安科技大学,2013. WANG Ya-lin.Algorithm of crowd density estimation based on gray level co-occurrence matrix[D].Xi'an:Xi'an University of Science and Technology,2013.
[14]陈 鸿.重点视频监控区域的人群密度估计改进算法的研究与实现[D].南京:南京邮电大学,2013. CHEN Hong.The research and implementation of the improve the crowded density estimation in important area[D].Nanjing:Nanjing University of Posts and Telecommunications,2013.
[15]Antoni B.Chan,Nuno Vasconcelos.Counting people with low-level featuresand bayesian regression[J].IEEE Transactions on Image Processing,2012,4(4):21-26.

备注/Memo

备注/Memo:
基金项目:国家自然科学基金项目(61302133); 陕西省工业攻关计划项目(2012K06-16); 西安科技大学博士启动金资助项目(2014QDJ066) 通讯作者:吴冬梅(1964-),女,浙江义乌人,教授,E-mail:wdm562@163.com
更新日期/Last Update: 2015-09-15