一种贝叶斯和支持向量机相结合的相关反馈策略

相关反馈技术在提高图像检索性能方面发挥着重要作用,但图像检索过程中的相关反馈存在反馈次数过多,反馈效果不够理想等问题。为解决上述问题,提出一种贝叶斯和支持向量机相结合的反馈算法。实现方法是:用贝叶斯分类器对图像库进行分类,达到压缩图像库的目的,然后用支持向量机分类器对压缩之后的图像库进行分类,并反馈最终结果。研究结果表明,与支持向量机和贝叶斯算法相比,在很少的反馈次数下,该方法明显提高了反馈效果。     

A practical formulation for estimating the extreme vector excursion of a floating structure

In the design of a moored floating structure, the traditional approach is to assess the extreme surge and sway responses separately. This paper highlights the importance of considering the extreme lateral excursion, based on the vector sum of the surge and sway offsets, and an approximate formula is developed for this purpose. The method allows for first- and second-order platform motions, as well as mean offsets. The extreme vector excursion is expressed as a simple function of the maximum surge and sway responses, and the correlation coefficient between the two motions. The proposed approximation is benchmarked against more rigorous methods, including time domain simulation. It is found that provided the estimates for the surge and sway extremes are precise, the formula itself does not induce significant errors in the extreme vector excursion. Moreover, it is observed that the prediction tends to be conservative. The practicality of the proposed formulation makes it amenable for incorporation into an existing design code or analysis procedure.     

支持向量机方法在单站降水预报中的应用探讨

将武汉天空云量预报的81个预报因子运用到该站中等以上强度的降水预报中,基于SVM方法进行了交叉验证和预报试验。结果表明用81个预报因子建立的5～9月和全样本的降水预报模型有较好稳定性、且对降水都有正的预报技巧。因此天空云量的预报因子可以用来做降水的预报因子,同时也证明了这些预报因子在天空云量和降水预报中是协调的。SVM方法为天空云量和降水的预报提供了客观参考依据。     

TanDEM-X multiparametric data features in sea ice classification over the Baltic sea

In this study, we assess the potential of X-band Interferometric Synthetic Aperture Radar imagery for automated classification of sea ice over the Baltic Sea. A bistatic SAR scene acquired by the TanDEM-X mission over the Bothnian Bay in March of 2012 was used in the analysis. Backscatter intensity, interferometric coherence magnitude, and interferometric phase have been used as informative features in several classification experiments. Various combinations of classification features were evaluated using Maximum likelihood (ML), Random Forests (RF) and Support Vector Machine (SVM) classifiers to achieve the best possible discrimination between open water and several sea ice types (undeformed ice, ridged ice, moderately deformed ice, brash ice, thick level ice, and new ice). Adding interferometric phase and coherence-magnitude to backscatter-intensity resulted in improved overall classification per- formance compared to using only backscatter-intensity. The RF algorithm appeared to be slightly superior to SVM and ML due to higher overall accuracies, however, at the expense of somewhat longer processing time. The best overall accuracy (OA) for three methodologies were achieved using combination of all tested features were 71.56, 72.93, and 72.91％ for ML, RF and SVM classifiers, respectively. Compared to OAs of 62.28, 66.51, and 63.05％ using only backscatter intensity, this indicates strong benefit of SAR interferometry in discriminating different types of sea ice. In contrast to several earlier studies, we were particularly able to successfully discriminate open water and new ice classes.     

陕西韩城地震台金属摆地倾斜资料与周边地震活动相关性探讨

列举陕西韩城地震台地倾斜资料对应汾渭地震带上的山西和陕西地区地震出现的震兆反应,表现在地倾斜矢量图有打折、打结和倾斜速率明显改变等异常。对于2008年5月12日四川汶川地震,地倾斜在震前一、二年出现矢量图变化速率明显改变的特征,认为属于震前异常。     

����֧����������ˮ�µ���ƥ�䵼�������������ַ����о�

????????????????????????α?????????/???????????????????????÷?????????????α????????????????????????????????????????????????????????????????????????????????????????α?????????/???????????????????     

基于移动激光扫描点云特征图像和SVM的建筑物立面半自动提取方法

建筑物立面是城市地物的重要组成部分,而移动激光扫描是获取城市地物三维信息的重要手段之一。本文提出了一种基于移动激光扫描点云的建筑物立面半自动提取算法。该方法首先构建研究区水平网格;然后计算局部点云几何特征,并且将特征投影到水平网格生成点云特征图像;接着基于支持向量机（Support Vector Machine,SVM）对建筑物立面网格进行粗提取;最后使用网格属性（形状系数、网格面积、最大高程）对粗提取结果进行过滤,并将结果反投影到三维空间中得到精确的建筑物立面。以卡内基梅隆大学的移动激光扫描点云进行试验后表明,本算法能够较好地提取出建筑物立面,提取精度为84%,召回率为90%,数据修正后精度为88%,召回率为91%。通过与现有算法对比,本文提出的算法具有较高精度。     

BP神经网络和SVM在矿山环境评价中的应用分析

矿山环境的影响因素多样,定量评价过程易受人为因素干预。BP神经网络与SVM算法能够自动模拟各因子间的非线性关系。首次将其引入到矿山环境评价中,选取160个单元作为训练样本,以自然地理、基础地质、开发占地及地质环境等4个大类的14个变量指标为输入向量,以单元评价得分为输出向量,分别建立BP神经网络与SVM矿山环境评价模型。结果表明：两种模型均能满足矿山环境评价的精度要求;SVM模型收敛速度较BP神经网络快,MSE小于BP神经网络,更适合矿山环境评价工作;将定量模型应用于研究区,评价得分划分为4个级别,与定性评价结果一致,为矿山环境评价工作提供了新思路。     

Structural damage detection using empirical-mode decomposition and vector autoregressive moving average model

A method based on empirical-mode decomposition (EMD) and vector autoregressive moving average (VARMA) model is proposed for structural damage detection. The basic idea of the method is that the structural damages can be identified as the abrupt changes in energy distribution of structural responses at high frequencies. Using the time-varying VARMA model to represent the intrinsic mode functions (IMFs) obtained from the EMD of vibration signal, we define a damage index according to the VARMA coefficients. In the two examples given, the Imperial County Services Building and the Van Nuys hotel are used as the benchmark structures to verify the effectiveness and sensitivity of the damage index in real environments with the presence of actual noise. The analysis results show that the damage index can indicate the occurrence and relative severity of structural damages at multiple locations in an efficient manner. The damage index can also be potentially used for structural health monitoring, since it is based on the time-varying VARMA coefficients. Finally, some recommendations for future research are provided.     

不同降维策略下的高光谱影像多特征分类

由于物体表面的空间分布通常是富有规律且局部连续的,在高光谱影像分类中应充分利用其光谱和空间信息.本文在对高光谱影像立方体进行降维处理的基础上,提出了一种联合空域和谱域信息的高光谱影像高效分类方法.首先,分别选用主成分分析(Principal Component Analysis,PCA)和正交投影波段选择(Orthog...