基于局部块模型的复杂场景行为识别算法

周英姿; 王正勇; 卿粼波; 何小海

doi:10.3788/YJYXS20173209.0748

您当前的位置：

首页 >

文章列表页 >

基于局部块模型的复杂场景行为识别算法

图像处理 | 更新时间：2020-07-07

- 基于局部块模型的复杂场景行为识别算法
- Action recognition algorithm under complex scenes based on local part model
- 液晶与显示 2017年32卷第9期页码：748-754
- 作者机构：
  
  四川大学电子信息学院, 四川成都 610065
- 作者简介：
  
  [ "周英姿(1995-), 女, 陕西咸阳人, 硕士研究生, 研究方向:数字图像处理, 模式识别等。E-mail:zyzjoyce77@163.com" ]
  [ "何小海(1964-), 男, 四川绵阳人, 教授, 博士生导师, 研究方向:图像处理, 网络通信, 数字通信等。E-mail: hxh@scu.edu.cn" ]
- 基金信息：
  
  成都市科技惠民项目(2015-HM01-00293-SF);特殊环境机器人技术四川省重点实验室开放基金(14zxtk03);国家自然科学基金委员会和中国工程物理研究院联合基金(11176018);the Technology Program of Public Wellbeing of Chengdu(2015-HM01-00293-SF);the Open Fund of Robot Technology Used for Special Environment Key Laboratory of Sichuan Province(14zxtk03);National Natural Science Foundation of China(11176018)
- DOI：10.3788/YJYXS20173209.0748
  中图分类号： TP394.1;TH691.9
- 收稿日期：2017-04-14，
  
  录用日期：2017-6-8，
  
  纸质出版日期：2017-09-05
- 稿件说明：
移动端阅览
周英姿, 王正勇, 卿粼波, 等. 基于局部块模型的复杂场景行为识别算法[J]. 液晶与显示, 2017,32(9):748-754.

Ying-zi ZHOU, Zheng-yong WANG, Lin-bo QING, et al. Action recognition algorithm under complex scenes based on local part model[J]. Chinese journal of liquid crystals and displays, 2017, 32(9): 748-754.
周英姿, 王正勇, 卿粼波, 等. 基于局部块模型的复杂场景行为识别算法[J]. 液晶与显示, 2017,32(9):748-754. DOI： 10.3788/YJYXS20173209.0748.

Ying-zi ZHOU, Zheng-yong WANG, Lin-bo QING, et al. Action recognition algorithm under complex scenes based on local part model[J]. Chinese journal of liquid crystals and displays, 2017, 32(9): 748-754. DOI： 10.3788/YJYXS20173209.0748.

摘要

针对视觉词袋模型与局部块模型特征相结合的算法在真实的复杂场景中识别率不高的问题，本文提出一种基于局部块模型与特征预处理、特征泛化相结合的行为识别算法。本文算法在视觉词袋模型的基础上，采取局部块模型与随机采样相结合的方法提取特征，对特征做预处理，减小了数据冗余，消除了特征之间的相关性，并且使处理后的特征更接近原始视频特征，同时对编码后特征做泛化处理，避免过拟合现象。本文在HMDB51标准视频库上进行实验，结果表明本文算法较原算法识别率提高2.1%，较其他同类算法也有一定的提升，验证了该算法的有效性。本文算法对视频量大、背景复杂、真实场景的视频集具有较好的识别效果。

Abstract

In order to solve the accuracy of the action recognition system in complex scenes

an improved method based on local part model

feature pre-processing and feature generalization is proposed. The algorithm is based on the bag of the visual words. Local part model and random sampling are used to extract features. The redundance of the features and the relationship between the features are reduced by the pre-processed procedure. The features processed are closer to the original features

meanwhile the features are generalized to prevent the over fitting. To demonstrate the performance of the proposed method

experiments are carried on HMDB51 datasets. The results show that the algorithm has higher efficiency than the original in complex environment. Compared with other methods

the proposed algorithm shows more accurate. It has a better recognition effect on video sets with large volume

complex background and real scene.

关键词

Keywords

references

KJELLSTRÖM H, ROMERO J, KRAGI D. Visual object-action recognition:inferring object affordances from human demonstration[J]. Computer Vision and Image Understanding, 2011, 115(1):81-90.

LAPTEV I. Onspace-time interest points[J]. International Journal of Computer Vision, 2005, 64(2/3):107-123.

ELSHOURBAGY M, HEMAYED E, FAYEK M. Enhanced bag of words using multilevelk-means for human activity recognition[J]. Egyptian Informatics Journal, 2016, 17(2):227-237.

SHI F, LAGANIERE R, PETRIU E. Gradient boundary histograms for action recognition[C]// Proceedings of 2015 IEEE Winter Conference on Applications of Computer Vision ( WACV ). Waikoloa, HI:IEEE, 2014:1107-1114.

刘富, 刘惠影, 高雷, 等.基于手指融合特征和粒子群优化的手形识别[J].光学精密工程, 2015, 23(6):1774-1782.

LIU F, LIU H Y, GAO L, et al. Hand shape recognition based on fusion features of fingers and particle swarm optimization[J]. Optics and Precision Engineering, 2015, 23(6):1774-1782. (in Chinese)

赵春阳, 赵怀慈.多模态鲁棒的局部特征描述符[J].光学精密工程, 2015, 23(5):1474-1483.

ZHAOC Y, ZHAO H C. Multimodality robust local feature descriptors[J]. Optics and Precision Engineering, 2015, 23(5):1474-1483. (in Chinese)

WANGH, SCHMID C. Action recognition with improved trajectories[C]// Proceedings of 2013 IEEE International Conference on Computer Vision ( ICCV ). Sydney, NSW, Australia:IEEE, 2013:3551-3558.

刘智, 黄江涛, 冯欣.构建多尺度深度卷积神经网络行为识别模型[J].光学精密工程, 2017, 25(3):799-805.

LIU Z, HUANG J T, FENG X. Action recognition model construction based on multi-scale deep convolution neural network[J]. Optics and Precision Engineering, 2017, 25(3):799-805. (in Chinese)

SHI F, PETRIU E M, CORDEIRO A. Human action recognition from local part model[C]// Proceedings of 2011 IEEE International Workshop on Haptic Audio Visual Environments and Games ( HAVE ). Hebei:IEEE, 2011:35-38.

SHI F, PETRIU E, LAGANIÈRE R. Sampling strategies for real-time action recognition[C]// Proceedings of 2013 IEEE Conference on Computer Vision and Pattern Recognition ( CVPR ). Portland, America:IEEE, 2013:2595-2602.

PENG X J, WANG L M, WANG X X, et al. Bag of visual words and fusion methods for action recognition:comprehensive study and good practice[J]. Computer Vision and Image Understanding, 2016, 150:109-125.

JORDAN M, KLEINBERG J, SCHOLKOPFB. Pattern Recognition and Machine Learning[M]. Germany:Springer, 2007.

FELZENSZWALB P, MCALLESTER D, RAMANAN D. A discriminatively trained, multiscale, deformable part model[C]// Proceedings of IEEE Conference on Computer Vision and Pattern Recognition . Anchorage, Alaska, America:IEEE, 2008:1-8.

KUMAR N, SETHI A. Fastlearning-based single image super-resolution[J]. IEEE Transactions on Multimedia, 2016, 18(8):1504-1515.

KANG S M, WILDES R P. Review of action recognition and detection methods[J]. arXiv :1610.06906, 2016.

KUEHNEH, JHUANG H, GARROTE E, et al . Hmdb:a large video database for human motion recognition[C]//P roceedings of 2011 IEEE International Conference on Computer Vision ( ICCV ). Barcelona, Spain:IEEE, 2011:2556-2563.

JIANG Y G, DAI Q, XUE X Y, et al. Trajectory-based modeling of human actions with motion reference points[C]// Proceedings of the 12 th European conference on Computer Vision . Florence, Italy:Springer, 2012:425-438.

王先中. 视频人体行为识别中的特征编码[D]. 上海: 上海交通大学, 2015. http://cdmd.cnki.com.cn/Article/CDMD-10248-1015806671.htm

WANG X Z. Feature coding in human action recognition[D]. Shanghai:Shanghai Jiao Tong University, 2015. (in Chinese)

LIU M Y, WANG R P, SHAN S G, et al. Learning mid-level words on riemannian manifold for action recognition[J]. arXiv :1511.04808, 2015.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据