基于多尺度特征深度学习的短临降水预报

雷达回波外推是短临降水预报的一种重要方法。针对雷达回波外推中随着外推时间的增加而出现回波演变信息丢失这一问题,本文提出一种多尺度特征融合的深度学习短临降水预报模型(以下简称为MSF2)。首先,采用多尺度的卷积核对网络的浅层信息进行特征提取,弥补单一特征检测带来的不足。其次,将不同维度的特征信息进行拼接及通道混洗,进一步增强特征图通道之间的信息流通和信息表达能力。最后,将特征图中的多尺度信息进行融合,从而有效保留不同尺度的特征信息。利用华南雷达回波拼图资料数据,在3种不同降水强度（5 mm/h、10 mm/h和25 mm/h）下进行降水预报研究,并与光流法和ConvLSTM两种主流算法进行了对比。结果显示,在3种不同降水强度条件下,MSF2在所有评价指标（命中率POD、临界成功指数CSI、误报率FAR）中表现最优,这表明引入多尺度机制能改善模型的特征提取能力。相比于目前主流的光流法和ConvLSTM,本文提出的模型对于短临降水预报具有较好的适用性和较高的预报精度,而且实现了业务化运行。     

基于深度学习和时空特征融合的海洋渔船密度预测方法

为了从海量渔船轨迹数据中挖掘隐含的信息和知识,进而为渔业行政主管部门的决策提供科学依据,本研究以AIS渔船轨迹数据为研究对象,提出了一种基于深度学习和面向时空特征融合的海洋渔船密度预测方法：首先,利用渔船轨迹数据集对渔船行驶区域进行网格划分;其次,筛选出渔船高密度区域进行研究,避免数据稀疏性问题;再次,根据渔船轨迹数据的时空分析,构建三维时空融合矩阵;最后,通过卷积循环神经网络模型捕获渔船分布的时间和空间特征,并利用卷积神经网络的堆叠加强对空间特征的学习。实验通过东海海域渔船真实轨迹数据进行具体测试,结果表明渔船密度预测值与真实值非常接近,平均绝对误差为4×10-4,模型较好地拟合了渔船密度分布特征,有效地提高了渔船捕捞热点预测的准确性和鲁棒性。     

基于极限学习机的GF-2影像分类

遥感图像分类是提取图像有效信息过程中重要的一部分,为了探寻最优的分类方法,许多机器学习算法逐步应用于遥感分类中。极限学习机（extreme learning machine,ELM）以其高效、快速和良好的泛化性能在模式识别领域得到广泛应用。本文采用训练速度快、运算量小的极限学习机算法与支持向量机（support vector machines,SVM）算法和最大似然法进行分类对比,对高分辨率遥感图像进行分类,分析极限学习机算法对于遥感图像分类的准确度等性能。选取吉林省长春市部分区域的GF-2遥感数据,将融合后的影像设置为原始数据,利用3种方法进行分类。研究结果表明,极限学习机算法分类图像总体分类精度达到85%以上,kappa系数达到0.718,与其他分类方法相比分类准确度较高,且极限学习机运行时间比支持向量机运行时间约短2 480 s,约为支持向量机运行时间的1/8,因此具有良好的性能和实用价值。     

基于年际增量法的卷积神经网络模型对中国夏季降水的预测研究

将前冬的500 hPa位势高度、向外长波辐射和海表温度的年际增量作为预测因子,建立基于卷积神经网络(Convolutional Neural Network, CNN)的非线性预测模型,对中国160个测站夏季降水展开预测研究,并与基于线性奇异值分解(Singular Value Decomposition, SVD)的预测模型进行效果对比。结果表明：CNN在1981—2020年的交叉检验中所回报的降水平均PS评分和距平相关系数(ACC)分别为74.33和0.12,比SVD高2.15和0.06,说明CNN比SVD在整体上对夏季降水具有更好的预测能力。其中,CNN对SVD预测较好年份的预测效果提升较为明显,对SVD预测较差的年份则改进不大。CNN对中国降水预测存在一定的系统性偏差,订正后CNN对拉尼娜年的降水预测改进较大。结果表明,基于年际增量法的CNN预测模型展示出较好的潜在应用价值。     

高分辨率遥感影像目标分类与识别研究进展

高分辨率遥感影像的目标分类与识别,是对地观测系统进行图像分析理解,以及自动目标识别系统提取目标信息的重要手段。本文综述了当前国内外在可见光、红外、合成孔径雷达和合成孔径声纳等遥感影像的目标分类与识别的关键技术和最新研究进展。首先,讨论了高分辨率遥感影像的目标分类与识别问题的主要研究层次和内容;其次,深入分析了高分辨率遥感影像目标分类与识别,在滤波降噪、特征提取、目标检测、场景分类、目标分类和目标识别的关键技术及其所存在的问题;最后,结合并行计算、神经计算和认知计算等技术,讨论了目标分类与识别的可行性方案。具体包括：（1）高性能并行计算在高分辨率遥感图像处理的主流技术,并给出了基于Hadoop+OpenMP+CUDA的高分辨率遥感影像混合并行处理架构;（2）深度学习对于提升目标分类和识别精度的应用前景,以及基于深度神经网络的多层次遥感影像目标识别方法;（3）认知计算在解决遥感影像大数据不确定性分析的模型与算法,并讨论了层次主题模型的多尺度遥感影像场景描述方案。此外,根据媒体神经认知计算的相关研究,探讨了遥感影像大数据的目标分类和识别的发展趋势和研究方向。     

Statistical models for shear strength of RC beam‐column joints using machine‐learning techniques

This paper proposes a new set of probabilistic joint shear strength models using the conventional multiple linear regression method, and advanced machine‐learning methods of multivariate adaptive regression splines (MARS) and symbolic regression (SR). In order to achieve high‐fidelity regression models with reduced model errors and bias, this study constructs extensive experimental databases for reinforced and unreinforced concrete joints by collecting existing beam‐column joint subassemblage tests from multiple sources. Various influential parameters that affect joint shear strength such as material properties, design parameters, and joint configuration are investigated through tests of statistical significance. After performing a set of regression analyses, the comparison of simulation results indicates that MARS approach is the best estimation method. Moreover, the accuracy of analytical predictions of the derived MARS model is compared with that of existing joint shear strength relationships. The comparison results show that the proposed model is more accurate compared to existing relationships. This joint shear strength prediction model can be readily implemented into joint response models for evaluation of earthquake performance and inelastic responses of building frames. Copyright © 2014 John Wiley & Sons, Ltd.     

移动互联网时代防震减灾科普宣传的问题及对策研究

在移动互联网蓬勃发展的信息技术背景下,我国防震减灾科普教育存在若干问题,如忽视受众的"数字化"特征,基于智能终端的宣传材料不足;宣传手段不能满足移动互联网时代信息化要求;宣传内容不够生动等,因此提出防震减灾科普知识泛在化宣传新模式,并根据泛在化宣传模式的特征,设计开发基于iOS系统的泛在化宣传资源,为防震减灾宣传工作提供基于移动设备的数字化资源支持。该研究对移动互联网时代地震科普宣传作了新的探讨与尝试,对于当前及未来全民普及防震减灾科学知识有积极的促进作用。     

Seafloor Classification of Multibeam Sonar Data Using Neural Network Approach

In this study, the self-organizing map (SOM), which is an unsupervised clustering algorithm, and a supervised proportional learning vector quantization (PLVQ), are employed to develop a combined method of seafloor classification using multibeam sonar backscatter data. The PLVQ is a generalized learning vector quantization based on the proportional learning law (PLL). The proposed method was evaluated in an area where there are four types of sediments. The results show that the performance of the proposed method is better than the SOM and a statistical classification method.     

Radial Basis Functional Link Nets Used as a Prospectivity Mapping Tool for Orogenic Gold Deposits Within the Central Lapland Greenstone Belt, Northern Fennoscandian Shield

Among the more popular spatial modeling techniques, artificial neural networks (ANN) are tools that can deal with non-linear relationships, can classify unknown data into categories by using known examples for training, and can deal with uncertainty; characteristics that provide new possibilities for data exploration. Radial basis functional link nets (RBFLN), a form of ANN, are applied to generate a series of prospectivity maps for orogenic gold deposits within the Paleoproterozoic Central Lapland Greenstone Belt, Northern Fennoscandian Shield, Finland, which is considered highly prospective yet clearly under explored. The supervised RBFLN performs better than previously applied statistical weights-of-evidence or conceptual fuzzy logic methods, and equal to logistic regression method, when applied to the same geophysical and geochemical data layers that are proxies for conceptual geological controls. By weighting the training feature vectors in terms of the size of the gold deposits, the classification of the neural network results provides an improved prediction of the distribution of the more important deposits/occurrences. Thus, ANN, more specifically RBFLN, potentially provide a better tool to other methodologies in the development of prospectivity maps for mineral deposits, hence aiding conceptual exploration.     

面向电网巡检的多旋翼无人机航测系统关键技术研究及应用

针对无人机电力巡检所面临的定位精度低、近距离巡检操控难度大、自动化程度低、人为操控事故风险高等问题,本文提出了利用多旋翼无人机RTK差分定位技术、复杂塔型飞行规则机器学习算法等无人机航测系统技术,研发面向电网巡检的无人机智能操作系统及控制终端,提升电网线路日常管理和应急指挥的规范性、便利性和经济性,为电网管理部门提供一种低成本、易操作、高效率、机动灵活的新型巡检模式。