基于宇宙计算的图形处理器算法实现

下一代观测望远镜将会产生数以亿计的星系测量数据值,这将导致使用中央处理器处理数据时效率低下、成本较高。为了解决这一问题,提出了基于宇宙计算的图形处理器算法。研究了两点式角相关函数以及孔径质量统计这两种宇宙学的计算方法,构建算法代码,并使用统一计算设备架构在图形处理器上实现了这两种算法;比较了算法在中央处理器和图形处理器上使用的运行速度。实验结果表明,与中央处理器相比,使用图形处理器的计算速度得到了显著提高。     

面向个性化云服务的动态信任模型

为更好地实践云计算为用户提供廉价按需服务的宗旨,满足服务请求者的个性化需求,提出一种面向个性化云服务的动态信任模型。基于细粒度服务思想定义个性化云服务,通过引入时间衰减因子和建立高效激励机制修正直接信任值,以灰色系统理论为基础计算实体间的评价相似度,并将评价相似度和推荐者的推荐可信度作为合成推荐信任值的重要因素,同时提出一种基于评价相似度的自信因子赋值方法,以提高合成综合信任值的准确性。实验结果表明,与GM-Trust模型及CCIDTM模型相比,该模型的交互成功率分别平均提高了4%和11%。     

最小相关性最大依赖度属性约简

在经典粗糙集中,基于重要度的决策表属性约简算法只考虑了决策属性与条件属性之间的依赖度,没有考虑约简中条件属性之间的相关性,由此求出的约简中可能依然包含冗余属性。针对这一问题,提出了一种改进算法,它利用最小相关性和最大依赖度准则求决策表属性约简。与基于重要度的决策表属性约简算法相比,本算法求出的约简包含的属性个数少、冗余小。实验结果显示,本算法优于基于重要度的决策表属性约简算法。     

基于非线性关联维特征提取的机械自动化监测系统

提取天然气压缩机故障状态下振动信号的故障特征是设计机械自动化检测系统的核心技术。提出一种基于非线性关联维特征提取的机械自动化监测系统设计方法,在故障诊断原理基础上,进行故障振动信号时间序列分析,设计故障振动信号相空间重构方法,改进了相空间重构最佳时延和嵌入维数参数计算的关键技术,通过提取关联维故障特征,在Simulink平台上设计了自动化监测系统。系统实验结果表明,该算法和系统能使各类故障状态下提取的关联维特征的标准差显著性降低,特征分布聚类能力明显提高,系统能有效检测各类故障,实现了机械设备的自动化监测,在自动化故障诊断仪表设计等领域具有较好的工程实践价值。     

Prediction of liquid film mass transfer coefficients in packed columns using liquid holdup

A unique characteristic linear dimension (d), defined as the cube root of the specific liquid holdup (h_sp) in the packed column, was used to correlate successfully the liquid film mass transfer coefficient k_La for gas absorption-desorption for sparingly soluble gases in liquids below loading. To produce this simple, dimensionless correlation, k_La data reported in literature were used, covering a wide range of physical properties of liquids, packings and operating conditions. This new approach showed operating holdup as an important factor in gas liquid mass transfer.     

基于半监督学习的一种图像检索方法

为了提高图像检索的质量, 提出了一种基于半监督学习的图像检索方法。该方法提取图像的颜色、形状、纹理特征, 计算得到已知类别样本的中心图像, 检索过程中利用中心图像自适应调整相关度计算函数, 然后根据与查询图像相关度的大小对图像进行检索和排序。实验结果表明：该方法较已有的基于内容的图像检索方法有更高的查准率, 同时, 由查准率—查全率曲线可知该方法能够达到很好的检索质量。     

Two-dimensional correlation relaxometry studies of cement pastes performed using a new one-sided NMR magnet

We present preliminary results of the first NMR T₁-T₂ two-dimensional relaxation correlation experiments performed using a one-sided NMR system in cement based materials. Two-dimensional correlation relaxometry has itself only recently been demonstrated in cement paste where it proved to be a particularly sensitive probe of pore-water dynamics providing direct information on exchange of water between the gel and capillary pore networks. Further to this we have observed differences in the structural development of a selection of cement pastes throughout the early stages of hydration and verified the theoretical frequency dependence of the ratio T₁ / T₂. When coupled with instrumentation developments in one-sided NMR magnets the way is opened to detailed, spatially resolved studies of the development of hydration and porosity in the surface layers (top 50 mm) of cementitious materials. A new magnet, suitable for such applications, is discussed.     

基于熵权灰色关联和D-S证据理论的威胁评估

为了更好地处理空中威胁目标的不确定性信息,提出了基于熵权灰色关联和D-S证据理论相结合的威胁评估方法.将熵理论应用于求解各指标权重,利用灰色关联法确定各指标的不确信度,进而得到各指标下不同目标的Mass函数,通过D-S证据理论对各Mass函数进行合成,根据置信函数大小对目标进行排序.仿真实验证明该方法是合理有效的.     

基于特征融合的加权SVM音频隐写分析算法

针对隐写分析中特征维数过高的问题,提出一种特征加权支持向量机音频隐写分析算法.利用特征相关性对原始特征进行优化选择,利用增益比率法计算特征权重,提出了改进特征加权支持向量机.与常用的C-SVM进行的对比实验表明,该方法能够有效提高检测率,降低时间复杂度.     

A minimum variance distortionless response filter based on the bifrequency spectrum for single-channel noise reduction

This paper deals with the problem of single-channel noise reduction in the short-time Fourier transform (STFT) domain. Many algorithms have been developed to solve this important problem, most of which generally assume that the STFT coefficients in different frequency bands are uncorrelated, so the noise reduction is achieved by applying a gain function to the STFT of the noisy speech in each frequency band. However, this assumption is not accurate and the STFT coefficients of speech signals between neighboring frequency bands are correlated in practice due to the use of small lengths of the fast Fourier transform (FFT) and overlap add/save techniques in implementation. This paper formulates the noise reduction problem by taking into account the interband correlation using the so-called bifrequency spectrum. Based on this formulation, a single-channel minimum variance distortionless response (MVDR) filter is derived, which is shown to be able to significantly improve the signal-to-noise ratio (SNR) and meanwhile maintain the desired speech not much distorted. Simulations are presented to justify the claimed merits of the developed MVDR filter.