一种海洋混合层深度的智能识别方法研究

文章提出了一种识别混合层深度的人工智能方法。该方法在温度(密度)与压强(或深度)间建立线性模型, 并且将其系数和方差做成一组表征廓线特征的统计量。初始时为模型设定一个主观的先验分布, 在一个自海表向下移动的窗口内通过贝叶斯链式法则和最小描述长度原理学习新数据, 得到系数均值的最大后验概率估计。用F-检验识别系数发生突变的位置, 以此确定混合层的存在性及其深度。通过2017年2月太平洋海域的地转海洋学实时观测阵(Array for Real-time Geostrophic Oceanography, ARGO)数据进行测试, 并且以质量因子(Quality Index, QI)值作为判断识别混合层深度结果准确性的依据, 发现该方法相比于梯度法、阈值法、混合法、相对变化法、最大角度法和最优线性插值法在识别结果上具备更大的QI值。表明该方法能够准确识别混合层深度。     

Indoor localization for pedestrians with real-time capability using multi-sensor smartphones

ABSTRACT

The localization of persons or objects usually refers to a position determined in a spatial reference system. Outdoors, this is usually accomplished with Global Navigation Satellite Systems (GNSS). However, the automatic positioning of people in GNSS-free environments, especially inside of buildings (indoors) poses a huge challenge. Indoors, satellite signals are attenuated, shielded or reflected by building components (e.g. walls or ceilings). For selected applications, the automatic indoor positioning is possible based on different technologies (e.g. WiFi, RFID, or UWB). However, a standard solution is still not available. Many indoor positioning systems are only suitable for specific applications or are deployed under certain conditions, e.g. additional infrastructures or sensor technologies. Smartphones, as popular cost-effective multi-sensor systems, is a promising indoor localization platform for the mass-market and is increasingly coming into focus. Today’s devices are equipped with a variety of sensors that can be used for indoor positioning. In this contribution, an approach to smartphone-based pedestrian indoor localization is presented. The novelty of this approach refers to a holistic, real-time pedestrian localization inside of buildings based on multi-sensor smartphones and easy-to-install local positioning systems. For this purpose, the barometric altitude is estimated in order to derive the floor on which the user is located. The 2D position is determined subsequently using the principle of pedestrian dead reckoning based on user's movements extracted from the smartphone sensors. In order to minimize the strong error accumulation in the localization caused by various sensor errors, additional information is integrated into the position estimation. The building model is used to identify permissible (e.g. rooms, passageways) and impermissible (e.g. walls) building areas for the pedestrian. Several technologies contributing to higher precision and robustness are also included. For the fusion of different linear and non-linear data, an advanced algorithm based on the Sequential Monte Carlo method is presented.     

基于贝叶斯正则化BP神经网络的DEM趋势面逼近

趋势面从宏观上揭示了研究对象的特性,在各领域发挥着重要作用。BP神经网络可以对复杂系统进行无限逼近,进而进行预测。建立了基于贝叶斯正则化BP神经网络的数字高程模型趋势面,与二次多项式建立的数字高程模型趋势面进行比较分析,证明了该方法的可行性和有效性。     

基于语义相似度的海量遥感图像可视化研究

针对海量遥感图像如何有效的传达分类结果以实现有效的可视化问题,本文按照分类的语义标注结果的相似度并运用现有的信息可视化技术来实现图像的可视化。首先采用了贝叶斯网络学习的方法进行图像的自动分类标注,然后利用基于图像布局的多维标度算法(Multi-dimensional Scale)以及无需降维的Value and Relation(VaR)技术实现可视化。实验表明本文的方法能够填补图像低层视觉特征和高层语义之间鸿沟,对大量的图像在一个视图内进行有效的浏览,而不造成图像的混乱,并能实现高层次的图像分析。实验的可视化结果是十分有效的。     

Comparing spatially varying coefficient models: a case study examining violent crime rates and their relationships to alcohol outlets and illegal drug arrests

In this paper, we compare and contrast a Bayesian spatially varying coefficient process (SVCP) model with a geographically weighted regression (GWR) model for the estimation of the potentially spatially varying regression effects of alcohol outlets and illegal drug activity on violent crime in Houston, Texas. In addition, we focus on the inherent coefficient shrinkage properties of the Bayesian SVCP model as a way to address increased coefficient variance that follows from collinearity in GWR models. We outline the advantages of the Bayesian model in terms of reducing inflated coefficient variance, enhanced model flexibility, and more formal measuring of model uncertainty for prediction. We find spatially varying effects for alcohol outlets and drug violations, but the amount of variation depends on the type of model used. For the Bayesian model, this variation is controllable through the amount of prior influence placed on the variance of the coefficients. For example, the spatial pattern of coefficients is similar for the GWR and Bayesian models when a relatively large prior variance is used in the Bayesian model.      

基于Membrane MRF模型的SAR图像贝叶斯去斑

SAR图像可以看作是真实反映地物后向散射特性的无噪图像与相干斑噪声的乘积,通过贝叶斯估计从图像观测值估计出图像真值即可去除相干斑.而贝叶斯去斑的关键在于建立能与SAR图像特性相匹配的先验信息模型.用MembraneMRF模型对先验信息建模,克服了以往所用GMRF模型对参数估计十分敏感的问题,并通过对该模型邻域结构的自适应调整来分类处理处于匀质区域和含结构特征区域的像元,在有效抑制相干斑的同时较好地保持图像的结构特征.仿真和实际SAR图像数据的实验结果,验证了所提方法的有效性.     

基于贝叶斯网络的海洋生态环境状况评价——以山东省为例

在海洋开发利用活动日益增多,海洋环境保护压力日益增大的背景下,进行海洋生态环境状况科学评价方法的研究可以为开展海洋生态环境监测评估与海洋生态环境保护政策制定提供科学依据,促进改善海洋生态环境状况。本文首先基于贝叶斯网络基本理论,构建了海洋生态环境状况贝叶斯网络评价模型;再以山东省为例,充分融合专家经验知识与客观数据,从定性与定量两个角度评价了2013—2019年山东省近海生态环境状况,分析了各评价指标间关系及其对近海生态环境状况的影响程度。评价结果表明2013—2019年山东省近海生态环境状况保持一般状态,海洋环境质量与海洋生态灾害类指标对评价结果影响最大,山东省海洋环境保护工作初见成效。研究结果表明该评价模型能有效评价海洋生态环境状况,分析结果对于环境保护政策制定具有一定的借鉴意义。     

基于加权贝叶斯网络的海洋灾害风险评估

全球气候变化背景下,海洋灾害的群发性、难以预见性和灾害链效应日显突出,造成的损失逐年上升,开展海洋灾害的风险评估工作至关重要。针对海洋灾害评估中的不确定问题,本文首先基于风险理论剖析了海洋灾害风险的不确定性特征,构建了灾害评估指标体系;然后基于贝叶斯网络模型,提出针对不确定性灾害评估的风险贝叶斯网络,进而基于主客观定权,构建了加权贝叶斯网络评估模型;最后对我国沿海地区海洋灾害开展评估研究。实验表明,该评估模型有效实现海洋灾害的风险评估,具有实际可操作性。