基于贝叶斯决策的电传飞控系统状态监测与健康管理

针对飞控系统早期故障信号的微弱性和不易检测性,提出了一种基于贝叶斯决策的飞控系统状态监测和健康管理新算法;采用最大似然估计法对故障特征向量概率密度函数进行参数估计,监测飞控系统故障特征概率密度分布相对于正常状态的偏离,将不易检测的故障信号转化为容易观测到的偏离信息,实现飞控系统状态的健康评估;以某型电液伺服作动器为例,采用上述方法对其进行状态监测和健康管理,仿真结果证明了该方法的可行性和有效性。     

基于多尺度斑点检测的快速视盘定位方法

提出了一种利用多尺度斑点检测的快速视盘定位方法。首先通过斑点检测确定12个视盘的候选区域,然后综合利用候选点邻域内的外观和血管特性决策视盘真实位置。对五个公开的数据集1540幅图像进行了测试,定位准确率达到97.9%。算法可以在缩小后的图像上进行检测,因此检测效率高,在五个数据集中平均检测时间约0.2s,远快于目前绝大部分检测算法。实验表明,对正常和轻度病变图像视盘检测效果很好,但是对严重病变的图像以及部分低质量的图像中的视盘检测,准确率仍有待提高。     

基于信息增益率的WNB水下机器人故障分类

提高故障诊断能力对于确保水下机器人系统的稳定运行具有重要意义,故障分类是目前水下机器人故障诊断所面临的一个重要问题。针对水下机器人推进器系统数据特征,提出一种基于信息增益率的加权朴素贝叶斯故障分类算法。首先,计算故障训练样本的先验概率,将各属性的信息增益率作为权值;其次,构建基于增益率加权的朴素贝叶斯分类模型;然后,对检测的故障数据利用分类模型获取具有最大后验概率的故障模式,实现故障分类。与朴素贝叶斯算法和决策树算法相比,仿真实验结果表明基于信息增益率加权的朴素贝叶斯算法的分类成功率更高,能够有效地实现水下机器人的故障分类。     

基于改进Hu不变矩的图像篡改检测算法

提出了一种针对图像的复制-粘贴篡改进行检测的方法。算法以图像子带编码为基础,利用Hu不变矩和Zernike矩的不变量对篡改图像进行检测。对图像做子带编码处理,得到近似子带(LL)图像,对近似子带图像进行Hu不变矩和Zernike矩的特征提取,用来表征图像的信息,最后采用上述特征信息进行图像的相关匹配。实验中将所提算法与极余弦变换(PCT)算法以及图像Zernik矩算法进行比较,结果表明所提算法对图像的旋转、平移攻击具有更优质的鲁棒性,提高了检测准确率。     

凸优化耦合传感器模式噪声的图像伪造检测

现有的图像伪造检测算法主要是借助局部像素与恒虚警率来决策真伪,且忽略了源图像的强烈空间相关性,使算法鲁棒性不佳,难以检测微小尺寸伪造。对此,根据成像传感器的独特随机特性,设计传感器模式噪声检测思想;并提出了凸优化机制耦合传感器模式噪声的图像伪造检测算法。基于光响应非均匀性噪声,联合马尔可夫随机场与贝叶斯规则,设计传感器模式噪声;并构造最佳图像标记像素的先验概率模型;嵌入贝叶斯规则,代替恒虚警率,考虑源图像的强烈空间依赖性,联合整个图像像素,确定最大概率标记像素映射;设计凸优化机制,将图像伪造检测转换为凸问题,提高算法检测效率。并分析了不同伪造区域尺寸对算法检测的影响。仿真结果表明：与当前图像伪造检测算法相比,本文算法具备更好的接收机工作特征;以及更高的检测精度与检测效率。     

基于对称性评估的保偏光纤偏振轴检测技术

提出一种基于端面图像对称性特征的偏振轴检测算法。该算法利用图像极坐标变换来实现图像的对称性检测,使用金字塔搜索法快速寻找最优对称轴,确定偏振轴向。所提算法利用图像的全局性特征而不依赖于应力区边缘点位置实现精确定位,对图像清晰度要求显著降低。实验结果表明,所提算法具有较高的准确性和精确性(±0.1°),其稳健性大幅增强,且算法速度提升了近1.5倍。     

高分辨率合成孔径雷达图像的Gamma分布下最大后验概率降斑算法

针对传统的Gamma分布下最大后验概率降斑算法不能有效保留均匀区域的点目标, 不能有效保留弱边缘以及不能有效滤除强边缘区域的斑点等问题, 提出了基于第二类统计量的先验参数估计的高分辨率合成孔径雷达图像Gamma 分布下最大后验概率降斑算法. 使用Mellin卷积和斑点的乘性模型, Gamma先验分布的参数可由观察图像的前两阶对数累积量精确估计.所提算法具有解析的滤波输出, 便于实现.农田和城区的高分辨率合成孔径雷达图像的降斑实验表明, 与传统的Gamma分布下最大后验概率降斑算法相比, 所提算法既能有效保留均匀区域的点目标, 又能有效保留弱边缘, 还能有效滤除强边缘区域的斑点. 关键词： 高分辨率合成孔径雷达图像 Gamma分布下最大后验概率降斑算法 第二类统计量 对数累积量     

贝叶斯概率图像自动分割研究

探讨了一种新的图像自动分割的方法。提出应用高斯有限混合模型与期望－极大化算法对图像特征空间的数据进行聚类，采用信息理论准则（ITC）确定要分割的图像区域数目，用贝叶斯概率分割图像。整合这些技术可以实现图像自动分割，而且实验结果表明信息理论准则可以确定适当的区域数目。     

不确定海洋环境中基于主成分量的稳健检测器

传统的信号检测算法在不确定的海洋环境中性能出现下降。基于贝叶斯原理的最优检测算法可以实现对不确定海洋环境中信号的有效检测,但是其突出问题是计算量较大。本文提出了一种基于主成分量分析的稳健信号检测器,该检测器利用贝叶斯原理将环境先验信息引入到检测算法中,同时使用主成分量分析方法来降低运算量,实现了对信号的快速有效检测。分别使用标准失配海洋模型和海上实测数据进行了计算机仿真和实验验证,结果表明:(1)基于主成分量的稳健信号检测器检测性能达到最优贝叶斯检测器的效果。(2)本文方法在线运算速度是贝叶斯最优检测器的5^一8倍。(3)环境先验信息失配的情况下,扩大海洋环境参数模型的不确定度范围有助于提高检测性能。      

基于深度学习的合成孔径雷达图像去噪算法

针对合成孔径雷达在生成图像时产生的固有斑点噪声问题,提出了一种基于多尺度注意力级联卷积神经网络的去噪算法,用多尺度卷积网络与注意力机制来实现图像的特征提取,将级联网络结构作为主网络中的特征增强部分,并加入批量归一化来防止模型中出现过拟合。实验结果表明,所提算法相较于其他传统图像去噪算法,峰值信噪比、结构相似性分别平均提高了0.75 dB～14.45 dB和0.01～0.16,在图像熵上也优于其他算法,并且能较好地恢复图像中的细节信息。     

Efficient sampling for decision making in materials discovery

Accelerating materials discovery crucially relies on strategies that efficiently sample the search space to label a pool of unlabeled data. This is important if the available labeled data sets are relatively small compared to the unlabeled data pool. Active learning with efficient sampling methods provides the means to guide the decision making to minimize the number of experiments or iterations required to find targeted properties. We review here different sampling strategies and show how they are utilized within an active learning loop in materials science.     

A New Hybrid Possibilistic-Probabilistic Decision-Making Scheme for Classification

Uncertainty is at the heart of decision-making processes in most real-world applications. Uncertainty can be broadly categorized into two types: aleatory and epistemic. Aleatory uncertainty describes the variability in the physical system where sensors provide information (hard) of a probabilistic type. Epistemic uncertainty appears when the information is incomplete or vague such as judgments or human expert appreciations in linguistic form. Linguistic information (soft) typically introduces a possibilistic type of uncertainty. This paper is concerned with the problem of classification where the available information, concerning the observed features, may be of a probabilistic nature for some features, and of a possibilistic nature for some others. In this configuration, most encountered studies transform one of the two information types into the other form, and then apply either classical Bayesian-based or possibilistic-based decision-making criteria. In this paper, a new hybrid decision-making scheme is proposed for classification when hard and soft information sources are present. A new Possibilistic Maximum Likelihood (PML) criterion is introduced to improve classification rates compared to a classical approach using only information from hard sources. The proposed PML allows to jointly exploit both probabilistic and possibilistic sources within the same probabilistic decision-making framework, without imposing to convert the possibilistic sources into probabilistic ones, and vice versa.     

A Bayesian approach to seafloor classification using multi-beam echo-sounder backscatter data

Seafloor classification using acoustic remote sensing techniques is an attractive approach due to its high-coverage capabilities and limited costs. The multi-beam echo-sounder (MBES) system provides high-resolution bathymetry and backscatter information with 100% coverage. In this paper, we present a seafloor classification method that employs the MBES backscatter data. The method uses the averaged backscatter data per beam. It, therefore, is independent on the quality of the MBES calibration. Also, its performance is insensitive to seafloor type variation along the MBES swathe and corrections for the angular dependence of the backscatter are not needed. The method accounts for the ping-to-ping variability of the backscatter intensity. It estimates both the number of seafloor types present in the survey area and the probability density function for the backscatter strength at a certain angle for each of the seafloor types. Application of the method to MBES backscatter data acquired in a well-known test area in the North Sea shows very good agreement with available ground truth. The method’s discriminatory performance for this area is demonstrated to be comparable to that of taking samples of the sediment. All seafloor types known to be present in the area are resolved for. Application of the method to the Stanton bank data set shows clearly separable areas that differ in seafloor composition.     

Bayesian Analysis of Exafs Data

In this paper we demonstrate the use of Bayesian analysis methods for the analysis of EXAFS data. We will discuss the physical parameters that may be estimated by the method and demonstrate the applicability of the method to Molybdenum coordination compounds.     

Information Theory for Agents in Artificial Intelligence,Psychology, and Economics

This review looks at some of the central relationships between artificial intelligence, psychology, and economics through the lens of information theory, specifically focusing on formal models of decision-theory. In doing so we look at a particular approach that each field has adopted and how information theory has informed the development of the ideas of each field. A key theme is expected utility theory, its connection to information theory, the Bayesian approach to decision-making and forms of (bounded) rationality. What emerges from this review is a broadly unified formal perspective derived from three very different starting points that reflect the unique principles of each field. Each of the three approaches reviewed can, in principle at least, be implemented in a computational model in such a way that, with sufficient computational power, they could be compared with human abilities in complex tasks. However, a central critique that can be applied to all three approaches was first put forward by Savage in The Foundations of Statistics and recently brought to the fore by the economist Binmore: Bayesian approaches to decision-making work in what Savage called ‘small worlds’ but cannot work in ‘large worlds’. This point, in various different guises, is central to some of the current debates about the power of artificial intelligence and its relationship to human-like learning and decision-making. Recent work on artificial intelligence has gone some way to bridging this gap but significant questions remain to be answered in all three fields in order to make progress in producing realistic models of human decision-making in the real world in which we live in.     

Baseline Roll Removal in NMR Spectra Using Bayesian Analysis

In this paper we demonstrate the use of Bayesian analysis methods for the removal of baseline roll from NMR spectra. Baseline roll occurs in NMR spectra due to experimental artifacts in the first few channels of the acquired time-domain signal. Bayesian analysis is used to predict improved estimates of these values which are substituted into the experimental time-domain signal prior to Fourier transformation. Although applied to NMR spectra, the method is general and may be applied to baseline correction in most spectroscopic techniques.     

Bayesian Deconvolution

Bayesian probability has been applied to the problem of estimating the true spectral lineshape from an experimental peak with prior knowledge of the instrumental contribution. The results demonstrate that Bayesian deconvolution provides excellent estimates of the parameters defining the true spectral lineshape even from strongly overlapping or broadened peaks in the presence of significant noise levels.     

基于稀疏贝叶斯学习的复杂网络拓扑估计

提出了一种噪声环境下复杂网络拓扑估计方法, 仅利用含噪时间序列估计未知结构混沌系统的动力学方程和参数, 以及由混沌系统组成的复杂网络的拓扑结构、节点动力学方程、所有参数、 节点间耦合方向和耦合强度.通过采用动力学方程的统一形式, 将动力系统方程结构和参数估计看成线性回归问题的系数估计, 该估计问题利用贝叶斯压缩传感的信号重建算法求解, 含噪信号的模型重建使用相关向量机方法,即通过稀疏贝叶斯学习求解稀疏欠定线性方程得到上面提到的可估计对象.以单个Lorenz系统及由200个 Lorenz系统组成的无标度网络为例说明方法的有效性. 仿真结果表明,提出的方法对噪声有很强的鲁棒性,收敛速度快,稳态误差极小, 克服了最小二乘估计方法收敛速度慢、 稳态误差大以及压缩传感估计方法对噪声鲁棒性不强的缺点.     

Designing neural networks that process mean values of random variables

We develop a class of neural networks derived from probabilistic models posed in the form of Bayesian networks. Making biologically and technically plausible assumptions about the nature of the probabilistic models to be represented in the networks, we derive neural networks exhibiting standard dynamics that require no training to determine the synaptic weights, that perform accurate calculation of the mean values of the relevant random variables, that can pool multiple sources of evidence, and that deal appropriately with ambivalent, inconsistent, or contradictory evidence.