变形分析的Bayes估计与检验

由于观测条件的变化，观测值方差将产生随机波动。本文根据Ｂａｙｅｓ估计理论，推导了线性模型待估参数的分布函数及观测值方差估计公式，并将其用于位移检验。计算结果表明，在同样的置信度下，本方法更符合实际。文中还对参数估值的统计性质进行了讨论。     

用贝叶斯方法推算设计波高

采用长期概率预报提供波浪要素设计值是海洋工程中不可缺少的内容之一，文中介绍了一种基於贝叶斯原理的新计算方法，它较传统统计方法有明显的优越性，不仅可以通过补充新信息不断修正原有的估算结果，而且可以避免传统方法带来诸多不确定性，文中以北海中部百年一遇设计波高为例进行了计算。     

半参数平差模型估计量的精度评定

将半参数估计理论引入测量数据处理理论中,利用Blight和Ott提出的思想,用多项式函数来表示半参数平差模型中的非参数分量,从而得到了半参数平差模型中参数分量和非参数分量的Bayes估计量,通过理论证明,半参数模型参数分量的Bayes估计为通常意义下高斯马尔可夫模型参数估计值与参数期望的加权平均,是一致、渐近无偏和渐近有效估计量,并且其方差小于参数模型中参数估计量的方差。在一定情况下推广了平差理论,具有一定的理论价值。     

人工智能在冰雹识别及临近预报中的初步应用

基于广东10部S波段多普勒天气雷达的三维拼图资料,利用机器学习技术开发了一种冰雹识别和临近预报的人工智能算法。算法设计时以雷达回波反射率的垂直和水平扫描数据为基础训练集,将冰雹云的雷达反射率扫描数据作为正样本,将其他雷达反射率扫描数据作为负样本,通过贝叶斯分类法对正、负样本数据集进行机器学习,训练人工智能识别冰雹云内在规律的能力。训练时以广东省2008-2013和2015-2016年的数据作为训练集,使用了2014年广东省12次冰雹过程的数据做检验。对比检验的结果表明,人工智能法比传统的概念模型法击中率高9个百分点。研究结果表明了人工智能对冰雹这类非线性强天气过程具有较强的识别能力。      

Fast Nearest Neighbor Classification Methods for Multispectral Imagery

Nearest neighbor classifiers have not been widely used by remote sensing practitioners. The lack of acceptance of these classifiers may be partially due to their notoriously slow speed of execution which makes them impractical for the classification of mega-pixel images. However, training data reduction, distance measure optimization, and neighbor searching algorithms based on the modified k-d tree can speed nearest neighbor classification substantially.     

Bayes判别分析法与地震短临预测

建立了一套完整的Bayes判别分析方法,并提出了明确的衡量综合预测结果所冒风险大小的风险代价比Kdn该方法可用于不同时间尺度的地震综合预测。选取云南地区1994～1996年的水氡观测资料,采用X2统计检验法识别前兆异常,利用Bayes判别分析方法对云南地区的强震活动性进行了内符检验和外推预测。在风险代价比Kdn取4的情况下,内符检验的R值可达0．54,外推时空占有率0．07,获得了满意的效果。     

Appraisal of Seismic Hazard Parameters for the Seismic Regions of the East Circum-Pacific Belt Inferred from a Bayesian Approach

The assessment of seismic hazard parameters is important in the seismically active regions. A straightforward approach is considered for the statistical estimation of the maximum values of earthquake hazard parameters. The Bayesian estimator is suggested and emphasis is given to the evaluation of the maximum possible M_max (regional) magnitude in a future time interval T. This approach allows the uncertainty of earthquake magnitude to be accounted for. Seismic hazard parameters like the -value which is the slope of the magnitude-frequency law (where, b = log_e) and the intensity (rate) of seismic activity and their uncertainties are also estimated. The quantiles of functions of distributions of true and apparent magnitude on a given time interval [0, T] are evaluated, as well. Two main assumptions are adopted for the method:(1) earthquake occurrence is Poissonian and(2) the magnitude-frequency law is of Gutenberg-Richter type with a cutoff maximum value of magnitude. It is needless to say the seismic catalog used must have a large number of events. This requirement leads to the estimation of the parameters referred to some of the most seismically active regions of the world, e.g., Chile, Peru-Equador-South Colombia,Central America and Mexico, which belong to the east part of the circum-Pacific belt.     

组合分类器及其在高光谱影像分类中的应用

多分类器组合问题一直是模式识别和机器学习领域的研究重点。由于各分类器的分类结果有着互补信息,因而一般来说,组合分类方法总有着较单个分类器精度更高的分类结果。将组合分类方法引入到高光谱影像的分类过程中,并针对该类影像的特点,提出了混合组合策略。实验表明,将组合分类器用于高光谱影像的分类,可以得到较为理想的分类结果。     

Detecting Arctic snow and ice cover with FY-1D global data

Chinese meteorological satellite FY-1D can obtain global data from four spectral channels which include visible channel(0.58-0.68 μm) and infrared channels(0.84-0.89 μm,10.3-11.3 μm,11.5-12.5 μm).2366 snow and ice samples,2024 cloud samples,1602 land samples and 1648 water samples were selected randomly from Arctic imageries.Land and water can be detected by spectral features.Snow-ice and cloud can be classified by textural features.The classifier is Bayes classifier.By synthesizing five d ays classifying result of Arctic snow and ice cover area,complete Arctic snow and ice cover area can be obtained.The result agrees with NOAA/NESDIS IMS products up to 70%.