文件名称:activity-recognition--based-on-hmm
介绍说明－－下载内容均来自于网络，请自行研究使用
一种HMM可以呈现为最简单的动态贝叶斯网络。隐马尔可夫模型背后的数学是由LEBaum和他的同事开发的。它与早期由RuslanL.Stratonovich提出的最优非线性滤波问题息息相关，他是第一个提出前后过程这个概念的。
在简单的马尔可夫模型（如马尔可夫链），所述状态是直接可见的观察者，因此状态转移概率是唯一的参数。在隐马尔可夫模型中，状态是不直接可见的，但输出依赖于该状态下，是可见的。每个状态通过可能的输出记号有了可能的概率分布。因此，通过一个HMM产生标记序列提供了有关状态的一些序列的信息。注意，“隐藏”指的是，该模型经其传递的状态序列，而不是模型的参数；即使这些参数是精确已知的，我们仍把该模型称为一个“隐藏”的马尔可夫模型。隐马尔可夫模型以它在时间上的模式识别所知，如语音，手写，手势识别，词类的标记，乐谱，局部放电和生物信息学应用。
隐马尔可夫模型可以被认为是一个概括的混合模型中的隐藏变量（或变量），它控制的混合成分被选择为每个观察，通过马尔可夫过程而不是相互独立相关。最近，隐马尔可夫模型已推广到两两马尔可夫模型和三重态马尔可夫模型，允许更复杂的数据结构的考虑和非平稳数据建模。-The HMM is a statistical approach in which the underlying model is a stochastic Markovian
process that is not observable (i.e., hidden) whic h can be observed through other processes that
produce the sequence of observed (emitted) features. In our HMM we let the hidden nodes represent
activities. The observable nodes re present combinations of the features described earlier. The
probabilistic relationships between hidden nodes and observable nodes and the probabilistic transition
between hidden nodes are estimated by the relative fr equency with which these relationships occur in
the sample data. An example HMM for three of the activities is shown in Figure 3. Given an input
sequence of sensor events, our algorithm finds the mo st likely sequence of hidden states, or activities,
which could have generated the observed event sequence. We use the Viterbi algorithm to
identify this sequence of hidden states.
在简单的马尔可夫模型（如马尔可夫链），所述状态是直接可见的观察者，因此状态转移概率是唯一的参数。在隐马尔可夫模型中，状态是不直接可见的，但输出依赖于该状态下，是可见的。每个状态通过可能的输出记号有了可能的概率分布。因此，通过一个HMM产生标记序列提供了有关状态的一些序列的信息。注意，“隐藏”指的是，该模型经其传递的状态序列，而不是模型的参数；即使这些参数是精确已知的，我们仍把该模型称为一个“隐藏”的马尔可夫模型。隐马尔可夫模型以它在时间上的模式识别所知，如语音，手写，手势识别，词类的标记，乐谱，局部放电和生物信息学应用。
隐马尔可夫模型可以被认为是一个概括的混合模型中的隐藏变量（或变量），它控制的混合成分被选择为每个观察，通过马尔可夫过程而不是相互独立相关。最近，隐马尔可夫模型已推广到两两马尔可夫模型和三重态马尔可夫模型，允许更复杂的数据结构的考虑和非平稳数据建模。-The HMM is a statistical approach in which the underlying model is a stochastic Markovian
process that is not observable (i.e., hidden) whic h can be observed through other processes that
produce the sequence of observed (emitted) features. In our HMM we let the hidden nodes represent
activities. The observable nodes re present combinations of the features described earlier. The
probabilistic relationships between hidden nodes and observable nodes and the probabilistic transition
between hidden nodes are estimated by the relative fr equency with which these relationships occur in
the sample data. An example HMM for three of the activities is shown in Figure 3. Given an input
sequence of sensor events, our algorithm finds the mo st likely sequence of hidden states, or activities,
which could have generated the observed event sequence. We use the Viterbi algorithm to
identify this sequence of hidden states.
(系统自动生成,下载前可以参看下载内容)
下载文件列表
src
...\Makefile
...\ar.c
...\ar.h
...\crf.c
...\crf.h
...\hmm.c
...\hmm.h
...\lbfgs.c
...\lbfgs.h
...\nb.c
...\nb.h
...\nb.o