###
DOI:
工程科学与技术:2016,48(Z2):121-126
←前一篇   |   后一篇→
本文二维码信息
码上扫一扫!
改进的基于局部块模型的行为识别算法(研究生论坛)
(四川大学电子信息学院)
An Improved Action Recognition Algorithm Based on Local Part Model
(College of Electronics and Information Engineering, Sichuan University, Chengdu)
摘要
图/表
参考文献
相似文献
附件
本文已被:浏览 1655次   下载 190
投稿时间:2015-09-02    修订日期:2016-04-21
中文摘要: 针对基于局部时空特征与特征词袋模型相结合的算法中计算效率低、识别率不高的问题,提出了一种基于局部块模型与特征数据预处理结合的行为识别算法。该算法基于特征词袋模型,采用局部块模型提取特征,针对局部块模型算法中处理数据维度高,相关性强导致的识别率低的问题,将多变量的复杂问题简化为低维空间的简单问题,对其数据处理过程进行了改进,同时优化了局部块模型中的帧采样过程。与原算法相比计算效率与识别率都有较大提升,较其他同类算法也具有一定优势。两个通用视频库上的实验证明了算法的有效性。
Abstract:In order to solve action recognition system’s efficiency and accuracy in complex environment, an improved method based on local part model and feature pre-processing was proposed. The algorithm based on the bag of features, and local part model was used to extract features. The features were pre-processed through Principal Component Analysis methods, While the local part model of frames in the sampling process was optimized. Experiments carried on two benchmark datasets, respectively named UCF Sports and HMDB51. The results showed that this algorithm has higher efficiency and accuracy than the original in complex environment. Compared with other methods , the proposed algorithm showed more accurate and efficient.
文章编号:201500904     中图分类号:    文献标志码:
基金项目: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项目(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资助项目(11176018)
作者简介:
引用文本:
田晶晶,吴晓红.改进的基于局部块模型的行为识别算法(研究生论坛)[J].工程科学与技术,2016,48(Z2):121-126.
tianjingjing,wuxiaohong.An Improved Action Recognition Algorithm Based on Local Part Model[J].Advanced Engineering Sciences,2016,48(Z2):121-126.