[1]沈世斌,谢 非,牛友臣,等.基于混合高斯模型优化的运动人体跟踪方法[J].南京师范大学学报(工程技术版),2019,(01):051.[doi:10.3969/j.issn.1672-1292.2019.01.007]
 Shen Shibin,Xie Fei,Niu Youchen,et al.A Moving Human Body Tracking Method Based onOptimized Gaussian Mixture Model[J].Journal of Nanjing Normal University(Engineering and Technology),2019,(01):051.[doi:10.3969/j.issn.1672-1292.2019.01.007]
点击复制

基于混合高斯模型优化的运动人体跟踪方法
分享到:

南京师范大学学报(工程技术版)[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2019年01期
页码:
051
栏目:
信息工程
出版日期:
2019-03-30

文章信息/Info

Title:
A Moving Human Body Tracking Method Based onOptimized Gaussian Mixture Model
文章编号:
1672-1292(2019)01-0051-07
作者:
沈世斌1谢 非12牛友臣3王天洋12钟港林12谷全琪1
(1.南京师范大学电气与自动化工程学院,江苏 南京 210042)(2.南京师范大学江苏省三维打印装备与制造重点实验室,江苏 南京 210042)(3.中国东方航空江苏有限公司,江苏 南京 211113)
Author(s):
Shen Shibin1Xie Fei12Niu Youchen3Wang Tianyang12Zhong Ganglin12Gu Quanqi1
(1.School of Electrical and Automation Engineering,Nanjing Normal University,Nanjing 210042,China)(2.Jiangsu Key Laboratory of 3D Printing Equipment and Manufacturing,Nanjing Normal University,Nanjing 210042,China)(3.China Eastern Airlines Jiangsu Limited Company,Nanjing 211113,China)
关键词:
目标跟踪混合高斯模型光流法Camshift算法
Keywords:
target trackingGaussian mixture modeloptical flow methodCamshift algorithm
分类号:
TP391.4
DOI:
10.3969/j.issn.1672-1292.2019.01.007
文献标志码:
A
摘要:
复杂背景下运动人体目标的自动检测与跟踪效果常易受环境光线变化的干扰. 面向变光线环境下运动人体检测与跟踪,提出一种基于混合高斯模型优化的Camshift检测跟踪算法,首先采用混合高斯模型进行前景建模,将外界扰动作为背景信息进行处理; 然后进行色彩空间转换并计算反向投影值,进一步利用Meanshift迭代定位运动目标; 最后,通过更新混合高斯模型及后续帧的处理保持人体目标的有效检测及跟踪. 实验结果表明,该方法相较于传统的光流方法及Camshift算法,可更好地适应环境光线变化及枝叶晃动影响,较好地获取运动目标前景信息,提高运动人体目标的检测及跟踪精度.
Abstract:
The automatic detection and tracking effects of moving human target are always susceptible to the change of ambient light under the complex backgrounds. Aiming at the moving human target detection and tracking in variable lighting environments,the paper proposes a detection and tracking algorithm based on Camshift optimized by Gaussian mixture model. Firstly,the Gaussian mixture model is used to build the foreground model,and the external interference is settled as background information. Secondly,the color space is converted and backprojection values are calculated,and Meanshift iterative method is further used to locate the moving targets. Finally,the update of Gaussian mixture model and processing of subsequent frames can keep an effective detection and tracking of the moving human body. The experiment results show that the proposed algorithm has a better adaptive ability to the changing light of the ambient environment and the shaking of the foliage compared with traditional optical flow and Camshift methods. Besides,this algorithm can improve the detection and tracking accuracy of moving human body,and can better extract the foreground information of moving target.

参考文献/References:

[1] MOHANAD B,OTHMAN O K,KYAW K H,et al. Automated daily human activity recognition for video surveillance using neural network[C]//IEEE International Conference on Smart Instrumentation,Measurement and Applications. Putrajaya,Malaysia,2017.
[2]CHEN X,QING L B,HE X H,et al. From eyes to face synthesis:a new approach for human-centered smart surveillance[J]. IEEE access,2018,6(1):14567-14575.
[3]陈娟,应骏,王健,等. 一种基于帧间差与图像分割的运动目标分割方法[J]. 上海师范大学学报(自然科学版),2017,46(2):242-246.
CHEN J,YING J,WANG J,et al. A method of moving object segmentation based on inter frame difference and graph cuts[J]. Journal of Shanghai normal university(natural sciences edition),2017,46(2):242-246.(in Chinese)
[4]刘玉婷,席庆彪,刘慧霞. 一种基于帧间差的快速UAV运动目标检测方法[J]. 计算机仿真,2014,31(7):36-40.
LIU Y T,XI Q B,LIU H X. A method for UAV moving target detection based on inter frame difference[J]. Computer simulation,2014,31(7):36-40.(in Chinese)
[5]张水发,张文生,丁欢,等. 融合光流速度与背景建模的目标检测方法[J]. 中国图象图形学报,2011,16(2):236-243.
ZHANG S F,ZHANG W S,DING H,et al. Background modeling and object detecting based on optical flow velocity field[J]. Journal of image and graphics,2011,16(2):236-243.(in Chinese)
[6]毕国玲,续志军,陈涛,等. 基于随机聚类的复杂背景建模与前景检测算法[J]. 物理学报,2015,64(15):701-712.
BI G L,XU Z J,CHEN T,et al. Complex background model and foreground detection based on random aggregation[J]. Acta physica sinica,2015,64(15):701-712.(in Chinese)
[7]TOBIAS S,VOLKER E,THOMAS S.Robust local optical flow for feature tracking[J]. IEEE access,2018,6(1):1377-1387.
[8]GE Z D,CHANG F L,LIU H B.Multi-target tracking based on Kalman filtering and optical flow histogram[C]//Chinese Automation Congress. Jinan,China,2017.
[9]KEREM S,ANDREA R,RAFFAELE C,et al. FPGA-based hardware implementation of real-time optical flow calculation[J]. IEEE transactions on circuits and systems for video technology,2018,28(1):206-216.
[10]LIU Z,CHAN S C,WANG C,et al. Multi-view articulated human body tracking with textured deformable mesh model[C]//IEEE International Symposium on Circuits and Systems. Lisbon,Portugal,2015.
[11]XU Y H,YE M,TIAN Z H,et al. Locally adaptive combining colour and depth for human body contour tracking using level set method[J]. IET computer vision,2014,8(4):316-328.
[12]LIU Z,LIN Z C,WEI X,et al. A new model-based method for multi-view human body tracking and its application to view transfer in image-based rendering[J]. IEEE transactions on multimedia,2018,20(6):1321-1334.
[13]SANJAY S,DAYANG R B,SUZIAH B S,et al. A study of stochastic algorithms for 3D articulated human body tracking[C]//IEEE Second International Conference on Image Information Processing. Shimla,India,2013.
[14]GUPTA M,BEHERA L,VENKATESH K S,et al. A robust visual human detection approach with UKF-based motion tracking for a mobile robot[J]. IEEE systems journal,2015,9(4):1363-1375.
[15]LEE K H,HWANG J N,OKOPAL G,et al. Ground-moving-platform-based human tracking using visual SLAM and constrained multiple kernels[J]. IEEE transactions on intelligent transportation systems,2016,17(12):3602-3612.
[16]APIDET B,NATTHA J,HIROSHI S.A system for detection and tracking of human movements using RSSI signals[J]. IEEE sensors journal,2018,18(6):2531-2544.
[17]LI D,JIANG D H,BAO R,et al. Moving object tracking method based on improved Lucas-Kanade sparse optical flow algorithm[C]//International Smart Cities Conference. Wuxi,China,2017.
[18]SEONG H S,RHEE C E,LEE H J.A novel hardware architecture of the Lucas-Kanade optical flow for reduced frame memory access[J]. IEEE transactions on circuits and systems for video technology,2016,26(6):1187-1199.
[19]XIU C B,CHAI Z H,HE H Y,et al. Improved Camshift tracking method based on the edge suppression[C]//29th Chinese Control and Decision Conference. Chongqing,China,2017.

相似文献/References:

[1]王 丹,刘 怀.基于改进混合高斯模型的背景提取与更新[J].南京师范大学学报(工程技术版),2015,15(02):060.
 Wang Dan,Liu Huai.Background Extraction and Updating Based on Improved Gaussion Mixture Model[J].Journal of Nanjing Normal University(Engineering and Technology),2015,15(01):060.
[2]郭 青,柯 炜,唐万春,等.一种视距与非视距混合环境下RSS定位方法[J].南京师范大学学报(工程技术版),2017,17(04):071.[doi:10.3969/j.issn.1672-1292.2017.04.012]
 Guo Qing,Ke Wei,Tang Wanchun,et al.Research on the Positioning Technology of RSS in the MixedEnvironment of the Line of Sight and Non-line of Sight[J].Journal of Nanjing Normal University(Engineering and Technology),2017,17(01):071.[doi:10.3969/j.issn.1672-1292.2017.04.012]

备注/Memo

备注/Memo:
收稿日期:2018-04-27.
基金项目:国家自然科学基金(61601228)、江苏省自然科学基金(BK20161021)、江苏省高校自然科学基金(17KJB510031)、江苏省三维打印装备与制造重点实验室项目(BM2013006)资助开放课题(3DL201607).
通讯联系人:谢非,博士,副教授,研究方向:机器视觉与图像处理、机器学习与模式识别. E-mail:xiefei@njnu.edu.cn
更新日期/Last Update: 2019-03-30