杨世强 罗晓宇 乔丹 柳培蕾 李德信

摘 要：针对现有动作识别中对连续动作识别研究较少且单一算法对连续动作识别效果较差的问题，提出在单个动作建模的基础上，采用滑动窗口法和动态规划法结合，实现连续动作的分割与识别。首先，采用深度置信网络和隐马尔可夫结合的模型DBN-HMM对单个动作建模;其次，运用所训练动作模型的对数似然值和滑动窗口法对连续动作进行评分估计，实现初始分割点的检测;然后，采用动态规划对分割点位置进行优化并对单个动作进行识别。在公开动作数据库MSR Action3D上进行连续动作分割与识别测试，结果表明基于滑动窗口的动态规划能够优化分割点的选取，进而提高识别精度，能够用于连续动作识别。

关键词：隐马尔可夫模型;动作分割;动作识别;滑动窗口;动态规划

中图分类号： TP391.4

文献标志码：A

Abstract： Concerning the fact that there are few researches on continuous action recognition in the field of action recognition and single algorithms have poor effect on continuous action recognition， a segmentation and recognition method of continuous actions was proposed based on single motion modeling by combining sliding window method and dynamic programming method. Firstly， the single action model was constructed based on the Deep Belief Network and Hidden Markov Model （DBN-HMM）. Secondly， the logarithmic likelihood value of the trained action model and the sliding window method were used to estimate the score of the continous action， detecting the initial segmentation points. Thirdly， the dynamic programming method was used to optimize the location of the segmentation points and identify the single action. Finally， the testing experiments of continuous action segmentation and recognition were conducted with an open action database MSR Action3D. The experimental results show that the dynamic programming based on sliding window can optimize the selection of segmentation points to improve the recognition accuracy， which can be used to recognize continuous action.

Key words： Hidden Markov Model （HMM）; action segmentation; action recognition; sliding window; dynamic programming

0 引言

人體动作识别是近年来诸多邻域研究的热点[1]， 如视频监控[2]、人机交互[3]等领域。随着人口老龄化，服务机器人将在未来的日常生活中发挥重要作用，观察和反映人类行动将成为服务机器人的基本技能[4]。动作识别逐渐应用到人们生活和工作的各个方面，具有深远的应用价值。

动作行为一般是以连续动作的形式来体现，包含多个单一动作，行为识别时根据分割与识别的前后关系，可分为直接分割和间接分割。直接分割是先根据简单的参数大小变化确定分割边界，然后识别分割好的片段，如白栋天等[5]根据关节速度、关节角度的变化对动作序列进行初始分割，该方法较为简单快速，但对于较复杂的连续动作分割误差较大。……