应用在线随机森林投票的动作识别

王世刚; 鲁奉军; 赵文婷; 赵晓琳; 卢洋

doi:10.3788/OPE.20162408.2010

您当前的位置：

首页 >

文章列表页 >

应用在线随机森林投票的动作识别

信息科学 | 更新时间：2020-07-07

- 应用在线随机森林投票的动作识别
- Action recognition based on on-line random forest voting
- Editorial Office of Optics and Precision Engineeri 2016年24卷第8期页码：2010-2017
- 作者机构：
  
  吉林大学通信工程学院, 吉林长春 130012
- 作者简介：
  
  [ "王世刚(1962-)，男，吉林长春人，教授，博士生导师，1983年于东北大学获得学士学位，1997年于吉林工业大学获得硕士学位，2001年于吉林大学获得博士学位，研究方向为数字图象处理技术及应用。E-mail：wangshigang@vip.sina.com" ]
  鲁奉军(1991-)，男，吉林德惠人，硕士研究生，2013年于吉林大学获得学士学位，主要从事数字图像处理、机器学习等方面的研究。E-mail：lufengjun2012@sina.com E-mail：lufengjun2012@sina.com
- 基金信息：
  
  教育部博士学科点专项科研基金资助项目(No.20120061110091);吉林省科技发展计划资助项目(No.20150204006GX);长春市科技局资助项目(No.14KG007)
- DOI：10.3788/OPE.20162408.2010
  中图分类号： TP391.4
- 收稿日期：2015-11-09，
  
  录用日期：2015-12-14，
  
  纸质出版日期：2016-08-25
- 稿件说明：
移动端阅览
王世刚, 鲁奉军, 赵文婷, 等. 应用在线随机森林投票的动作识别[J]. Editorial Office of Optics and Precision Engineeri, 2016,24(8):2010-2017.

Shi-gang WANG, Feng-jun LU, Wen-ting ZHAO, et al. Action recognition based on on-line random forest voting[J]. Optics and precision engineering, 2016, 24(8): 2010-2017.
王世刚, 鲁奉军, 赵文婷, 等. 应用在线随机森林投票的动作识别[J]. Editorial Office of Optics and Precision Engineeri, 2016,24(8):2010-2017. DOI： 10.3788/OPE.20162408.2010.

Shi-gang WANG, Feng-jun LU, Wen-ting ZHAO, et al. Action recognition based on on-line random forest voting[J]. Optics and precision engineering, 2016, 24(8): 2010-2017. DOI： 10.3788/OPE.20162408.2010.

摘要

提出了基于在线随机森林投票识别人物动作类别的方法。建立了在线随机森林投票模型。通过在线训练和在线检测两部分进行了算法研究，提高了检测人物动作类别的准确率。基于人物动作在时间和空间上有重要信息，该方法首先通过提取图像立体块的lab色彩空间值、一阶差分、二阶差分以及大位移光流特征值在线训练随机森林；训练结束后，形成强分类器，利用分类器对检测图像进行投票，生成动作空间图；最后，在动作空间图中寻求最大值，判断检测图像的动作类别。验证结果表明在低分辨的视频图像中，本方法能够确定人物的动作类别，对Weizmann数据库和KTH数据库的识别率分别为97.3%和89.5%，对UCF sports数据库的识别率为79.2%，动作识别准确率有所提高。该方法增加了光流能量场特征表述，将原始投票理论拓展至三维空间，并且采用向下采样的方式更新结点信息，能够判断人物动作类别，为智能视频技术提供了有效的补充信息。

Abstract

An action recognition method for people is proposed based on on-line random forest voting to judge the action classification. The on-line random forest voting model is established and its algorithms are researched through the two parts consisting of on-line training and on-line detection to improve the precision of the action classfication. As people action shows important information in both space and time

the method firstly trains the random forests in line by extracting 3D image features containing a lab color space

the first order difference

the second order difference and displacement optical flow. After training

a strong classier is formed. Then

the classifier is used to vote for detection images to produce an action space map. Finally

by seeking the maximum in the map

the category of action in the detection images is complemented. Experimental results indicate that the method determines the category of people action in the low resolution video images. The accurate rates of the Weizmann data

the KTH data and the UCF sport data are 97.3%

89.5%

and 79.2%

respectively. These results show that the accuracy of action recognition is improved. Moreover

the model proposed adds the feature representation of light flow energy field

expands the traditional forest voting theory to a 3D space

and uses to update information. It improves the stability and the reliability and will be of potential application in the intelligent video surveillances.

关键词

Keywords

references

SCHULDT C, APTEV I, CAPUTO B. Recognizing human actions: a local SVM approach[C].IEEE, Proceedings of the 17th International Conference on Pattern Recognition, Cambridge, the United Kingdom, 2004: 32-36.

DOLLAR P, ABAUD V R, COTTRELL G, et al.. Behavior recognition via sparse spatio-temporal features [C]. Visual Surveillance and Performance Evaluation of Tracking and Surveillance, Beijing, P.R. China. 2005: 65-72.

付朝霞, 王黎明. 基于时空兴趣点的人体行为识别[J]. 微电子学与计算机, 2013, 30(8): 28-35.

FU ZH X, WANG L M. Human action recognition based on space-time interest point [J].Microelectronics & Computer, 2013, 30(8): 28-35. (in Chinese)

王博, 李燕. 视频序列中的时空兴趣点检测及其自适应分析[J]. 计算机技术与发展, 2014, 24(4): 49-56.

WANG B, LI Y. Space-time interest points detection in video sequence and its adaptive analysis [J]. Computer Technology and Development, 2014, 24(4): 49-56. (in Chinese)

王世刚, 孙爱朦, 赵文婷, 等. 基于时空兴趣点的单人行为及交互行为识别[J]. 吉林大学学报(工学版), 2015, 45(1): 304-308.

WANG SH G, SUN A M, ZHAO W T, et al.. Single and interactive human behavior recognition algorithm based on spatio-temporal interest point [J]. Journal of Jilin University (Engineering and Technology Edition), 2015, 45(1): 304-308. (in Chinese)

鲍一丹, 陈纳, 何勇, 等. 近红外高光谱成像技术快速鉴别国产咖啡豆品种[J]. 光学精密工程, 2015, 23(2): 349-355.

BO Y D, CHEN N, HE Y, et al.. Rapid identification of coffee bean variety by near infrared hyperspectral imaging technology [J]. Opt. Precision Eng., 2015, 23(2): 349-355. (in Chinese)

蔡加欣, 冯国灿, 汤鑫, 等. 基于局部轮廓和随机森林的人体行为识别[J]. 光学学报, 2014, 34(10): 1015006(1-10).

CAI J X, TANG G C, TANG X, et al.. Human action recognition based on local image contour and random forest[J]. Acta Optoca Sinica, 2014, 34(10): 1015006(1-10).(in Chinese)

BROX T, MALIK J. Large displacement optical flow: descriptor matching in variational motion estimation [J].IEEE, Transactons on Pattern Analysis and Machine Intelligence, 2011, 33(3): 500-513.

徐富元, 顾国华, 陈钱, 等. 转动红外探测器下地面远距离运动目标检测方法[J]. 红外与激光工程, 2014, 43(4): 1080-1086.

XU F Y, GU G H, CHEN Q, et al.. Ground target detection method on rotating infrared detector[J]. Infrared and Laser Engineering, 2014, 43(4): 1080-1086. (in Chinese)

向涛, 李涛, 李旭东, 等. 基于随机森林的层次行人检测算法[J]. 计算机应用研究, 2015, 32(7): 2196-2199.

XIANG T, LI T, LI X D, et al.. Random forests for hierarchical pedestrian detection [J].Application Research of Computers, 2015, 32(7): 2196-2199. (in Chinese)

屠大维, 江济良. 改进的光流运动图像分析方法及其应用 [J]. 光学精密工程, 2011, 19(5): 1159-1164.

TU D W, JIANG J L. Improved algorithm for motion image analysis based on optical flow and its application [J].Opt. Precision Eng, 2011, 19(5): 1159-1164. (in Chinese)

刘翔, 杨鑫, 王蕾. 改进的基于码本模型目标检测算法[J]. 液晶与显示, 2014, 29(6): 999-1002.

LI X, YANG X, WANG L. Object detection algorithm based on improved codebook model [J]. Chinese Journal of Liquid Crystals and Displays, 2014, 29(6): 999-1002. (in Chinese)

林培杰, 郑柏春, 陈志聪, 等. 面向多区域视频监控的运动目标检测系统[J]. 液晶与显示, 2015, 30(3): 484-491.

LI P J, ZHENG B CH, CHEN ZH C, et al.. System of multi-regions moving object detection in video surveillance [J]. Chinese Journal of Liquid Crystals and Displays, 2015, 30(3): 484-491. (in Chinese)

胡梦婕, 魏振忠, 张广军. 基于对象性测度估计和霍夫森林的目标检测方法[J]. 红外与激光工程, 2015, 44(6): 1936-1941.

HU M J, WEI ZH ZH, ZHANG G J. Object detection method based on objectness estimation and Hough forest[J]. Infrared and Laser Engineering, 2015, 44(6): 1936-1941. (in Chinese)

赵海, 陈星池, 王家亮, 等. 基于四轴飞行器的单目视觉避障算法[J]. 光学精密工程, 2014, 22(8): 2232-2241.

ZHAO H, CHEN X CH, WANG J L, et al.. Obstacle avoidance algorithm based on monocular vision for quad-rotor helicopter [J]. Opt. Precision Eng., 2014 22(8): 2232-2241 . (in Chinese)

JUERGEN G, VICTOR L. Class-specific Hough forests for object detection [C]. IEEE Transactions on Computer Vision and Pattern Recognition, Miami, the United States 2009: 1022-1029.

BLANK M, GORELICK L, SHECHTMAN E, et al.. Actions as space-time shapes[C]. IEEE International Conference on Computer Vision, Beijing, China, 2005.

SCHULDT C, LAPTEV I, CAPUTO B. Recognizing human actions: a local SVM approach[C]. Proceedings of the 17th International Conference on Pattern Recognition, Cambridge, UK, 2004.

RODRIGUEZ M, AHMED J, SHAH M. Action mach-a spatio-temporal maximum average correlation height filter for action recognition[C]. 2008 IEEE Conference on Computer Vision and Pattern Recognition, Anchorage, AK, 2008.

浏览量

532

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

利用几何特征和时序注意递归网络的动作识别

人体下肢应激微反应自动识别

基于Kinect数据主成分分析的人体动作识别