Deep Recurrent Neural Networks for Human Activity Recognition.
Deep Recurrent Neural Networks for Human Activity Recognition.
复制标题
DOI：
10.3390/s17112556
复制
发表时间：
2017-11-06
期刊：
影响因子：
--
通讯作者：
Pyun JY
中科院分区：
文献类型：
--
作者：
Murad A;Pyun JY
采用深度学习方法进行人体活动识别，可以有效地从身体佩戴传感器获取的原始输入序列中提取区分特征。虽然人类运动被编码在时间上的连续样本序列中，但典型的机器学习方法在不利用输入数据样本之间的时间相关性的情况下执行识别任务。卷积神经网络（CNN）通过在一维时间序列上使用卷积来捕获输入数据之间的依赖关系来解决这个问题。然而，卷积核的大小限制了数据样本之间依赖关系的捕获范围。因此，典型的模型是不适应广泛的活动识别配置，需要固定长度的输入窗口。在本文中，我们提出使用深度递归神经网络（DRNN）来构建能够捕获可变长度输入序列中的长程依赖关系的识别模型。我们提出了基于长短期记忆（LSTM）DRNN的单向、双向和级联架构，并评估了它们在各种基准数据集上的有效性。实验结果表明，我们提出的模型优于传统的机器学习方法，如支持向量机（SVM）和k-近邻（KNN）。此外，所提出的模型比其他深度学习技术（如深度信任网络（DBN）和CNN）具有更好的性能。
Adopting deep learning methods for human activity recognition has been effective in extracting discriminative features from raw input sequences acquired from body-worn sensors. Although human movements are encoded in a sequence of successive samples in time, typical machine learning methods perform recognition tasks without exploiting the temporal correlations between input data samples. Convolutional neural networks (CNNs) address this issue by using convolutions across a one-dimensional temporal sequence to capture dependencies among input data. However, the size of convolutional kernels restricts the captured range of dependencies between data samples. As a result, typical models are unadaptable to a wide range of activity-recognition configurations and require fixed-length input windows. In this paper, we propose the use of deep recurrent neural networks (DRNNs) for building recognition models that are capable of capturing long-range dependencies in variable-length input sequences. We present unidirectional, bidirectional, and cascaded architectures based on long short-term memory (LSTM) DRNNs and evaluate their effectiveness on miscellaneous benchmark datasets. Experimental results show that our proposed models outperform methods employing conventional machine learning, such as support vector machine (SVM) and k-nearest neighbors (KNN). Additionally, the proposed models yield better performance than other deep learning techniques, such as deep believe networks (DBNs) and CNNs.
登录
查看更多内容
影响因子：
5.1
作者：
Chavarriaga, Ricardo;Sagha, Hesam;Roggen, Daniel
通讯作者：
Roggen, Daniel
DOI：
10.1109/jbhi.2016.2633287
发表时间：
2017-01-01
影响因子：
7.7
作者：
Ravi, Daniele;Wong, Charence;Yang, Guang-Zhong
通讯作者：
Yang, Guang-Zhong
DOI：
10.1109/titb.2009.2036165
发表时间：
2010-03-01
影响因子：
--
作者：
Baechlin, Marc;Plotnik, Meir;Troester, Gerhard
通讯作者：
Troester, Gerhard
影响因子：
5.4
作者：
Schuster, M;Paliwal, KK
通讯作者：
Paliwal, KK
影响因子：
8.6
作者：
Sokolova, Marina;Lapalme, Guy
通讯作者：
Lapalme, Guy