Leveraging Transfer Learning for Spatio-Temporal Human Activity Recognition from Video Sequences

Human Activity Recognition (HAR) is an active research area due to its applications in pervasive computing, human-computer interaction, artificial intelligence, health care, and social sciences. Moreover, dynamic environments and anthropometric differences between individuals make it harder to recognize actions. This study focused on human activity in video sequences acquired with an RGB camera because of its vast range of real-world applications. It uses two-stream ConvNet to extract spatial and temporal information and proposes a fine-tuned deep neural network. Moreover, the transfer learning paradigm is adopted to extract varied and fixed frames while reusing object identification information. Six state-of-the-art pre-trained models are exploited to find the best model for spatial feature extraction. For temporal sequence, this study uses dense optical flow following the two-stream ConvNet and Bidirectional Long Short Term Memory (BiLSTM) to capture long-term dependencies. Two state-of-the-art datasets, UCF101 and HMDB51, are used for evaluation purposes. In addition, seven state-of-the-art optimizers are used to fine-tune the proposed network parameters. Furthermore, this study utilizes an ensemble mechanism to aggregate spatial-temporal features using a four-stream Convolutional Neural Network (CNN), where two streams use RGB data. In contrast, the other uses optical flow images. Finally, the proposed ensemble approach using max hard voting outperforms state-of-the-art methods with 96.30% and 90.07% accuracies on the UCF101 and HMDB51 datasets.     

基于上下文的概念语义相似度计算模型

针对概念语义相似度计算方法在信息检索中存在的漏检、误检等问题,提出一种基于上下文的计算本体内概念间语义相似度的模型。该模型分别从概念的父节点、子节点以及概念间路径权重3个角度进行计算,加权求和并求得语义相似度。实验结果表明,该模型的计算结果更加接近专家的经验值,为概念之间的语义关系提供了有效的量化。     

面向业务的E-维护知识供应模型

为使E-维护中的知识支持维护任务,提出面向业务的知识供应模型。采用维护知识概念本体规范任务描述、维护对象和知识功能,通过知识对象本体集成知识资源。将知识供应的需求和权限集成到业务过程中,构建维护知识供应信息模型,建立知识匹配方法控制知识供应。在某石化企业E-维护平台上的应用结果表明,该模型能提升企业的维护水平。     

时空数据异常探测方法

以“k倍标准差”准则为基础,提出一种专题属性双重偏离的时空异常检测方法,在每个要素的空间邻近域里采用“k倍标准差”准则探测各时刻的空间异常数据,在每个空间异常数据的时间邻近域中,再次使用该准则判断该要素是否为时序异常,并将所有空间和时间邻近域上均表现为异常的数据定义为时空异常。实验结果表明,该方法是有效可行的。     

老年人看护系统中常识推理的研究

根据老年人看护的需求,将常识推理运用到智能家庭老年人看护推理系统中,并将其与领域本体相结合,对系统实现的主要技术,即基于本体的常识表示与存储、规则的表示以及常识的推理进行探讨和构想,并通过典型应用实例验证了常识推理在老年人看护系统中的可行性、有效性及其重要性。     

Maximum likelihood analysis of algorithms and data structures

We present a new approach for an average-case analysis of algorithms and data structures that supports a non-uniform distribution of the inputs and is based on the maximum likelihood training of stochastic grammars. The approach is exemplified by an analysis of the expected size of binary tries as well as by three sorting algorithms and it is compared to the known results that were obtained by traditional techniques. Investigating traditional settings like the random permutation model, we rediscover well-known results formerly derived by pure analytic methods; changing to biased data yields original results. All but one step of our analysis can be automated on top of a computer-algebra system. Thus our new approach can reduce the effort required for an average-case analysis, allowing for the consideration of realistic input distributions with unknown distribution functions at the same time. As a by-product, our approach yields an easy way to generate random combinatorial objects according to various probability distributions.     

μC/OS-Ⅱ嵌入式操作系统实时性能分析与测量

以评价实时操作系统性能为目标,对时间参数的测量原理和方法进行了研究,以μC/OS-Ⅱ实时嵌入式操作系统为研究对象,分析其内核的实时机制及任务上下文切换时间、中断响应时间等实时性指标,给出了测量方法,并采用两种方法来获取时间参数,获得了一批嵌入式实时操作系统的数据,为实时性设计提供了参考。     

形式背景的交约简

从拓扑的角度研究形式背景的属性约简问题。对于非决策形式背景,提出基于集合交集的属性约简的新的定义,并得出了交协调集的判定定理,证明了交协调集与概念格协调集是等价的;最后,讨论了带有决策的形式背景的属性约简问题。     

结构上下文:一种新的物体类别描述符

局部描述符(如SIFT)方法能够将图像中关键点的局部表观信息作为图像的特征,具有旋转不变性、尺度变换不变性、仿射不变性等性质,被广泛应用于物体分类、物体识别、图像匹配等领域。但是,它存在一个重要缺陷:只能描述物体的局部特征,忽略了整个物体的构造,而这在表示物体时是非常重要的。设计了一个新的"结构上下文"局部描述符,通过当前关键点和其他关键点间的空间拓扑结构关系描述各个关键点的特征。实验证明这种描述符在描述相同物体种类时特别有效。     

MCSM:一种基于移动设备的上下文感知中间件

提出一种针对移动设备的上下文感知中间件MCSM(Mobile Context Sensing Middleware).MCSM可以动态地感知当前移动设备的内外部计算环境及用户活动等上下文信息,并以统一的接口将感知到的上下文数据提供给上层应用程序.为了克服外部传感器数据的高异构性,MCSM采用了一种基于XML的上下文数据描述语言CDDL.基于MCSM的原型系统表明,MCSM具有良好的运行效率和可扩展性,并可显著提高上下文应用程序的开发效率.