地球化学填图与地球化学块体的谱系分析——编号系统及谱系树图的自动绘制

中国的地球化学填图研究工作起步较早并发展成为在国际上具有明显优势的学科。在此基础上提出的地球化学块体理论通过一定的编号系统绘制谱系树图来追索某元素在块体内部逐步浓集成矿的轨迹。通过谱系树图可以发现不同级别的子块体之间的亲缘关系,研究它们在时间和空间上的分布规律,结合已发现矿床的信息,预测潜在矿床的分布及其储量,从而指导找矿工作。提出了一种全新的谱系编号系统并编制了在计算机上自动绘制谱系树图的软件,同时实现了与原有编号系统的一致,解决了长期以来谱系树图不能自动绘制的问题,为地球化学块体的研究提供了一个很好的工具。     

数据挖掘技术在水文预报中的应用及水文预报发展趋势研究

数据挖掘作为知识发现过程中的重要步骤,是从大型数据库中提取未知的、有价值的和可操作性的关系、模式和趋势用于决策支持的过程。我国目前在防洪领域存在着大量水文数据,如何充分有效地利用各种智能算法对这些数据进行分析与挖掘,以形成相应的水文预报模型进行准确的水文预报是防洪决策支持系统完善和发展的重要方面。本文首先全面地介绍数据挖掘的功能分类及以数据挖掘为基础的水文预报体系,然后对数据挖掘技术在水文预报中的应用进行了探讨和研究。     

A Comparison between a Modified Counter Propagation Network and an Extended Self-Organizing Map in Remotely Sensed Data Classification

A modified counter propagation network model and an extended self-organizing map model have the same three-layer network architecture while employing slightly different learning rules. Their network architecture comprises an input layer, a Kohonen layer and an output layer. The neurons between two neighboring layers are fully connected and the neighboring neurons within the Kohonen layer also have neighborhood connections. The modified counter propagation network model employs the Kohonen algorithm to train the Kohonen layer while using the Widrow–Hoff rule to train the output layer. However, the extended self-organizing map model applies a modified Kohonen’s learning rule to train both the Kohonen layer and the output layer. This paper compares the performances of these two models in supervised classification of remotely sensed data. The training results show that compared to the extended self-organizing map model, the modified counter propagation model has faster learning speed but larger output errors. The classification results indicate that the extended self-organizing map model has a faster classification speed and a much higher classification precision than the modified counter propagation model.     

Sensitivity analysis of seismic hazard for Western Liguria (North Western Italy): A first attempt towards the understanding and quantification of hazard uncertainty

The use of logic trees in probabilistic seismic hazard analyses often involves a large number of branches that reflect the uncertainty in the selection of different models and in the selection of the parameter values of each model. The sensitivity analysis, as proposed by Rabinowitz and Steinberg [Rabinowitz, N., Steinberg, D.M., 1991. Seismic hazard sensitivity analysis: a multi-parameter approach. Bull. Seismol. Soc. Am. 81, 796–817], is an efficient tool that allows the construction of logic trees focusing attention on the parameters that have greater impact on the hazard.In this paper the sensitivity analysis is performed in order to identify the parameters that have the largest influence on the Western Liguria (North Western Italy) seismic hazard. The analysis is conducted for six strategic sites following the multi-parameter approach developed by Rabinowitz and Steinberg [Rabinowitz, N., Steinberg, D.M., 1991. Seismic hazard sensitivity analysis: a multi-parameter approach. Bull. Seismol. Soc. Am. 81, 796–817] and accounts for both mean hazard values and hazard values corresponding to different percentiles (e.g., 16%-ile and 84%-ile). The results are assessed in terms of the expected PGA with a 10% probability of exceedance in 50 years for rock conditions and account for both the contribution from specific source zones using the Cornell approach [Cornell, C.A., 1968. Engineering seismic risk analysis. Bull. Seismol. Soc. Am. 58, 1583–1606] and the spatially smoothed seismicity [Frankel, A., 1995. Mapping seismic hazard in the Central and Eastern United States. Seismol. Res. Lett. 66, 8–21]. The influence of different procedures for calculating seismic hazard, seismic catalogues (epicentral parameters), source zone models, frequency–magnitude parameters, maximum earthquake magnitude values and attenuation relationships is considered. As a result, the sensitivity analysis allows us to identify the parameters with higher influence on the hazard. Only these parameters should be subjected to careful discussion or further research in order to reduce the uncertainty in the hazard while those with little or no effect can be excluded from subsequent logic-tree-based seismic hazard analyses.     

基于改进SOFM的矢量量化图像压缩

在介绍矢量量化和自组织特征映射神经网络的基础上，针对基于自组织特征映射神经网络的矢量量化算法，在初始码书生成、获胜神经元搜索策略以及调整获胜码字及其拓扑领域权值等方面进行改进。实验结果表明改进算法具有合理性和有效性。     

基于神经网络的遥感影像分类方法研究

本文在研究BP神经网络和模糊理论的基础上,提出了传统BP算法的一种改进方法和基于模糊系统的神经网络遥感影像分类方法。通过试验表明:基于模糊技术的神经网络分类方法要优于BP神经网络方法,取得了令人满意的效果。     

基于BP神经网络的三维激光测高数据拟合方法

针对三维激光测高拟合问题，本文采用BP神经网络进行拟合，该方法可以有效地消除或减弱数据中的系统误差影响，提高计算精度。实际算例证明，在数据无粗差情况下，BP神经网络的内符合与外符合精度优于传统的二次曲面拟合的精度。     

基于CART集成学习的城市不透水层百分比遥感估算

利用Landsat ETM^＋遥感数据，提出了一种基于CART集成学习的ISP遥感亚像元估算方法，将Boosting重采样技术引入CART分析中，用于提高ISP估算的精度。实验结果表明，该方法的ISP估算性能优于传统的单一CART学习算法，从ETM^＋影像中估算的ISP值与真实值之间的相关系数达到0．91，平均偏差为11．16％。     

基于自适应Kalman滤波的BP神经网络及其在高程拟合中的应用

针对传统BP神经网络容易发散、泛化能力差等问题，采用自适应Kalman滤波方法训练神经网络的连接权。与Kalman滤波训练连接权和传统的BP算法相比，该方法提高了BP神经网络计算精度，增强了泛化能力。实测数据的计算结果证明了该方法的可行性与有效性。     

Texture classification of Mediterranean land cover

Maximum likelihood (ML) and artificial neural network (ANN) classifiers were applied to three Landsat Thematic Mapper (TM) image sub-scenes (termed urban, agricultural and semi-natural) of Cukurova, Turkey. Inputs to the classifications comprised (i) spectral data and (ii) spectral data in combination with texture measures derived on a per-pixel basis. The texture measures used were: the standard deviation and variance and statistics derived from the co-occurrence matrix and the variogram. The addition of texture measures increased classification accuracy for the urban sub-scene but decreased classification accuracy for agricultural and semi-natural sub-scenes. Classification accuracy was dependent on the nature of the spatial variation in the image sub-scene and, in particular, the relation between the frequency of spatial variation and the spatial resolution of the imagery. For Mediterranean land, texture classification applied to Landsat TM imagery may be appropriate for the classification of urban areas only.