采用PHOG融合特征和多类别Adaboost分类器的行为识别

刘丽娜; 傅琪; 温加睿

doi:10.3788/OPE.20182611.2827

您当前的位置：

首页 >

文章列表页 >

采用PHOG融合特征和多类别Adaboost分类器的行为识别

信息科学 | 更新时间：2020-07-05

- 采用PHOG融合特征和多类别Adaboost分类器的行为识别
- Using PHOG fusion features and multi-class Adaboost classifier for human behavior recognition
- 光学精密工程 2018年26卷第11期页码：2827-2837
- 作者机构：
  
  1.上海大学机电工程与自动化学院, 上海 210072
  2.山东理工大学电气与电子工程学院, 山东淄博 255049
- 作者简介：
  
  [ "刘丽娜(1981-), 女, 山东邹平人, 博士研究生, 讲师, 2003年于青岛大学获得学士学位, 2006年于山东大学获得硕士学位, 主要从事模式识别与信息处理自动化、图像处理等的研究。E-mail: linaliu-126@163.com" ]
- 基金信息：
  
  国家自然科学基金资助项目(61671285);山东省自然科学基金资助项目(ZR2016FP04)
- DOI：10.3788/OPE.20182611.2827
  中图分类号： TP391.4
- 收稿日期：2017-12-11，
  
  录用日期：2018-2-11，
  
  纸质出版日期：2018-11-25
- 稿件说明：
移动端阅览
刘丽娜, 傅琪, 温加睿. 采用PHOG融合特征和多类别Adaboost分类器的行为识别[J]. 光学精密工程, 2018,26(11):2827-2837.

Li-na LIU, Qi FU, Jia-rui WEN. Using PHOG fusion features and multi-class Adaboost classifier for human behavior recognition[J]. Optics and precision engineering, 2018, 26(11): 2827-2837.
刘丽娜, 傅琪, 温加睿. 采用PHOG融合特征和多类别Adaboost分类器的行为识别[J]. 光学精密工程, 2018,26(11):2827-2837. DOI： 10.3788/OPE.20182611.2827.

Li-na LIU, Qi FU, Jia-rui WEN. Using PHOG fusion features and multi-class Adaboost classifier for human behavior recognition[J]. Optics and precision engineering, 2018, 26(11): 2827-2837. DOI： 10.3788/OPE.20182611.2827.

摘要

为了解决类能量图易受人体运动时间和位置移动等因素影响而难以有效描述动作细节特征的问题，本文提出了一种基于类能量图金字塔梯度直方图（PHOG）融合特征和多类别Adaboost分类器的人体行为识别方法。该方法首先对经过躯体配准的运动人体目标轮廓图像构造平均运动能量图（AMEI）和增强的运动能量图（EMEI），分别提取其分层梯度方向直方图（PHOG）特征并进行串联融合，作为一种多层次的行为特征描述；然后使用基于查找表的LUT-Real Adaboost算法设计多类别分类器，实现图像中人体行为动作的识别。实验结果显示其在典型的人体动作数据集DHA上的正确识别率达97.6%，高于其它采用单一特征描述和SVM等分类器的方法。表明该方法将整体与局部特征相结合，可以有效描述不同尺度下的动作细节特征，增强了人体行为特征的描述能力，提高了识别性能。

Abstract

In order to solve the problem that energy image species (EIS) are susceptible to human movement time and position shift

i.e.

it is difficult to describe the details of human behaviors

in this paper a method of human behavior recognition was present based on pyramid gradient histogram (PHOG) fusion features and a multi-class Adaboost classifier. This method first calculated the average motion energy image (AMEI) and the enhanced motion energy image (EMEI) of an object's silhouette images after human body registration

and then it extracted the PHOG features of AMEI and EMEI and series them together to form a kind of multi-level feature descriptor of human behavior. Finally

a look-up table-based real Adaboost (LUT-Real Adaboost) algorithm was utilized to realize human behavior recognition by designing a multi-class classifier. Experimental results show that the correct recognition rate in typical depth-included human action datasets is 97.6% by using this method

which is higher than that of other classifiers using single feature description and support vector machine. This reveals that

by combining global and local features

the proposed method can effectively describe the detailed active features of human behavior at different scales

enhance the description ability of human behavior characteristics

and improve recognition performance.

关键词

Keywords

references

GUO G D, LAI A. A survey on still image based human action recognition[J]. Pattern Recognition , 2014, 47(10), 3343-3361.

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

张国梁, 贾松敏, 张祥银, 等.采用自适应变异粒子群优化SVM的行为识别[J].光学精密工程, 2017, 25(6): 1669-1678.

ZHANG G L, JIA S M, ZHANG X Y, et al .. Action recognition based on adaptive mutation particle swarm optimization for SVM[J]. Opt. Precision Eng. , 2017, 25(6): 1669-1678.(in Chinese)

裴晓敏, 范慧杰, 唐延东.时空特征融合深度学习网络人体行为识别方法[J].红外与激光工程, 2018, 47(2): 0203007.

PEI X M, FAN H J, TANG Y D. Action recognition method of spatio-temporal feature fusion deep learning network[J]. Infrared and Laser Engineering , 2018, 47(2): 0203007.(in Chinese)

李庆辉, 李艾华, 崔智高, 等.结合限制密集轨迹与时空共生特征的行为识别[J].光学精密工程, 2018, 26(1): 230-237.

LI Q H, LI A H, CUI ZH G, et al .. Action recognition via restricted dense trajectories and spatio-temporal co-occurrence feature[J]. Opt. Precision Eng. , 2018, 26(1): 230-237.(in Chinese)

SAIKA S, TAKAHASHI S, TAKEUCHI M, et al .. Accuracy improvement in human detection using HOG features on train-mounted camera[C]. IEEE Global Conference on Consumer Electronics, IEEE , 2016: 1-2. https://www.researchgate.net/publication/312250277_Accuracy_improvement_in_human_detection_using_HOG_features_on_train-mounted_camera

GAO H L, CHEN W J. Image Classification Based on the Fusion of Complementary Features[J]. Journal of Beijing Institute of Technology , 2017, 26(2): 197-205.

申晓霞, 张桦, 高赞, 等.基于Kinect和金字塔特征的行为识别算法[J].光电子激光, 2014, 25(2): 357-363.

SHEN X X, ZHANG H, GAO Z, et al .. Human behavior recognition based on Kinect and pyramid features[J]. Journal of Optoelectronics·Laser , 2014, 25(2): 357-363.(in Chinese)

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.

LIU L N, WEN J R, MA S W, et al.. Human Behavior Recognition Method Based on Improved Energy Image Species and Pyramid HOG Feature [M]. Theory, Methodology, Tools and Applications for Modeling and Simulation of Complex Systems. Asia Sim 2016/ SCS Autumn Sim 2016, Part Ⅳ, CCIS 646, Singapore, 2016: 216-2224.

SHEN X X, ZHANG H, GAO Z, et al .. Human behavior recognition based on axonometric projections and PHOG feature[J]. Journal of Computational Information Systems , 2014, 10(8): 3455-3463.

WANG H, GAO J, TONG L, et al .. Facial recognition based on PHOG feature and sparse representation[C]. Proceedings of the 35th Chinese Control Conference. Chengdu, China, IEEE , 2016: 3869-3874. https://www.researchgate.net/publication/308498264_Facial_expression_recognition_based_on_PHOG_feature_and_sparse_representation

张昊.基于多尺度金字塔特征块提取HOG特征的新型人脸识别算法[D].长春: 吉林大学计算机科学与技术学院, 2017.

ZHANG H. A novel face recognition method using HOG features deriving from multi-layer pyramid feature blocks [D]. Changchun: School of Computer Science and Technology, Jilin University, 2017.(in Chinese)

杨冰, 王小华, 杨鑫, 等.基于HOG金字塔人脸识别方法[J].浙江大学学报(工学版), 2014, 48(9): 1564-1569.

YANG B, WANG X H, YANG X, et al .. Face recognition method based on HOG pyramid[J]. Journal of Zhejiang University (Engineering Science) , 2014, 48(9): 1564-1569. (in Chinese)

徐超, 高敏, 杨锁昌, 等.视觉注意机制下的粒子窗快速目标检测[J].光学精密工程, 2015, 23(11): 3227-3237.

XU CH, GAO M, YANG S CH, et al .. Visual attention mechanism-aided fast target detection by particle window[J]. Opt. Precision Eng. , 2015, 23(11): 3227-3237.(in Chinese)

HUANG W B, WANG K, YAN Y. Automatic detection method of blood vessel for color retina fundus images[J]. Opt. Precision Eng. , 2017, 25(5): 1378-1386.

CHEN C Y, ZHANG P Z, LUO L M. Face detection using real Adaboost on granular features[J]. Caai Transactions on Intelligent Systems , 2009.

HAN J, BHANU B. Individual recognition using gait energy image[J]. IEEE Transation on Pattern Analysis & Machine Intelligence , 2006, 28(2): 316-322.

LIN Y C, HU M C, CHENG W H, et al .. Human action recognition and retrieval using sole depth information[C]. The 20th ACM International conference on Multimedia(MM'12), Nara, Japan, ACM , 2012: 1053-1056. https://www.researchgate.net/publication/262363298_Human_action_recognition_and_retrieval_using_sole_depth_information

KLASER A, MARSZALEK M. A spatio-temporal descriptor based on 3d-gradients[C]. The 19th British Machine Vision Conference(BMVC 2008), Leeds, United Kingdom, British Machine Vision Association, BMVA , 2008, 9: 1-4. https://www.researchgate.net/publication/221259643_A_Spatio-Temporal_Descriptor_Based_on_3D-Gradients

SOLMAZ B, ASSARI S M, SHAH M. Classifying web videos using a global video descriptor[J]. Machine Vision and Applications , 2013, 24(7): 1473-1485.

林春丽, 王科俊, 李玥, 等.基于增强能量图和二维保局映射的行为分类算法[J].计算机应用, 2011, 31(3): 721-723.

杨丽召.基于多特征融合的行为识别算法研究[D].成都: 电子科技大学计算机科学与工程学院, 2013. http://cdmd.cnki.com.cn/Article/CDMD-10614-1013330291.htm

YANG L ZH. Aresearch of Behavior Recognition Algorithms Based on Multi-features Fusion [D]. Chengdou: School of Computer Science and Engineering, University of Electronic Science and Technology of China, 2013.(in Chinese)

欧阳寒, 范勇, 高琳, 等.基于归一化R变换分层模型的人体行为识别[J].计算机工程与设计, 2013, 34(6): 2170-2174.

OUYANG H, FAN Y, GAO L, et al .. Hierarchical human action recognition based on normalized R-transform[J]. Computer Engineering and Design , 2013, 34(6): 2170-2174.(in Chinese)

申晓霞, 张桦, 高赞, 等.一种鲁棒的基于深度数据的行为识别算法[J].光电子激光, 2013(8): 1613-1618.

SHEN X X, ZHANG H, GAO Z, et al .. A robust behavior recognition algorithm based on sole depth information[J]. Journal of Optoelectronics·Laser , 2013(8): 1613-1618.(in Chinese)

浏览量

149

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