Extending self-organizing maps for supervised classification of remotely sensed data

An extended self-organizing map for supervised classification is proposed in this paper. Unlike other traditional SOMs, the model has an input layer, a Kohonen layer, and an output layer. The number of neurons in the input layer depends on the dimensionality of input patterns. The number of neurons in the output layer equals the number of the desired classes. The number of neurons in the Kohonen layer may be a few to several thousands, which depends on the complexity of classification problems and the classification precision. Each training sample is expressed by a pair of vectors ： an input vector and a class codebook vector. When a training sample is input into the model, Kohonen＇s competitive learning rule is applied to selecting the winning neuron from the Kohouen layer and the weight coefficients connecting all the neurons in the input layer with both the winning neuron and its neighbors in the Kohonen layer are modified to be closer to the input vector, and those connecting all the neurons around the winning neuron within a certain diameter in the Kohonen layer with all the neurons in the output layer are adjusted to be closer to the class codebook vector. If the number of training sam- ples is sufficiently large and the learning epochs iterate enough times, the model will be able to serve as a supervised classifier. The model has been tentatively applied to the supervised classification of multispectral remotely sensed data. The author compared the performances of the extended SOM and BPN in remotely sensed data classification. The investigation manifests that the extended SOM is feasible for supervised classification.     

基于深度学习的赤潮发生预报方法研究

赤潮作为海洋灾害,对海洋渔业、生态、经济,以及人类生产、生活造成了严重影响。一直以来,赤潮受到研究者的广泛关注,但由于它的形成机制比较复杂,使得赤潮预报极具挑战性。针对赤潮预报的研究问题,本文收集了厦门海域赤潮发生前后的海洋监测数据,结合皮尔逊相关系数、散布矩阵、复相关系数方法,分析多环境因子与赤潮发生多要素的关联情况,重点采用基于深度学习的LSTM与CNN融合方法,挖掘环境因子的时序依赖,发现序列数据的局部特征,对赤潮发生进行预报。在厦门一号和厦门二号数据集中,本方法在预报未来12 h内的赤潮情况时,RMSE、MAE误差分别达到0.521 8、0.504 3。通过协同对比模型进一步确定赤潮发生的预报概率,在两个数据集上的最终预报准确率分别为67.58%和63.49%。本研究为赤潮的分析预报提供了探索经验,证明了将深度学习方法应用于赤潮预报的可行性。     

基于融合U-Net及ConvLSTM的海面高度异常预报方法研究

海面高度异常是反映海洋环境状况的主要变量之一。本文使用1993—2019年的融合月均海面高度异常数据,建立了基于深度学习的海面高度异常预测神经网络模型,提出了基于融合U型网络(U-Net)和卷积长短记忆网络(ConvLSTM)的中长期海面高度异常预报模型。在研究海域0.25°×0.25°的空间分辨率下,模型测试集预报结果的均方根误差和平均绝对误差分别为0.039 m和0.027 m,均优于全连接LSTM预报模型和ConvLSTM+CNN预报模型,为大中尺度的海面高度异常预报提供了新的方法。     

基于迁移学习方法的山东冷云冰晶特征分布研究

为分析冷云中冰晶的分布特征,揭示冰晶增长演变机制,根据冰晶形状和尺度特征分为8类并进行标注,同时标注1类隔断栏进行数据质量控制,将9类标签图像整合并建立图像集,利用迁移学习VGGNet16方法进行识别训练,经训练模型分类准确率达98%。将模型应用到秋季冷云冰晶特征研究中,选取3次积层混合云和3次层状云降水过程,分析冰晶形状在不同温度区间的占比及冰晶谱变化特征,结果表明,温度通过影响冰晶基面与棱面的比值来决定冰晶初始形状分布,相同温度区间积层混合云内球状冰晶和线型冰晶占比高于层状云,低于-12℃后各类冰晶占比相对固定;积层混合云内线型冰晶直径集中在300~800 μm,冰晶谱呈多峰分布,聚合体直径大于600 μm,冰晶谱首尾两端浓度相当,球状冰晶直径集中在120～300 μm,冰晶谱呈单调下降趋势。     

基于机器学习的格点气温预报订正方法

使用2017年9月至2021年3月国家级业务化运行的智能网格实况分析产品和欧洲中期天气预报中心全球模式(EC)产品,根据湖北省的地理分布特征构建6个分区,采用基于LightGBM机器学习算法建立的气温预报方法,生成湖北省0.05°×0.05°格点气温预报产品。利用2021年4—9月的预报产品和格点实况资料进行检验,结果表明:基于机器学习的气温预报方法(MLT)取得了较好的预报效果,其在0～72 h时效内优于中央气象台下发的气温精细化指导预报(SCMOC)和EC产品;MLT在山区的误差较平原大,但山区的订正幅度大于平原,日最高气温的订正幅度大于日最低气温的订正幅度;4—9月MLT、SCMOC、EC产品的平均绝对误差(MAE)日变化都呈现了白天偏高、夜间偏低、午后凸起的单峰特征,MLT的MAE值较SCMOC和EC产品的更低,并且在转折性天气中仍具有优势;站点检验与格点检验结论一致,基于格点建模的气温预报产品对站点预报同样得到了订正。机器学习在格点气温的模式订正方面可以作为一个行之有效的手段。     

遥感影像分析在新增建设用地督察中的应用

以卫星影像大数据和时空数据挖掘技术为基础,采用深度学习框架支撑,构建了面向新增建设用地遥感监测应用需求的深度学习训练平台,可智能发现自然资源督察线索,全面提高自然资源督察问题的发现能力.该技术可广泛服务于大范围、高频次的督察工作,为督察工作提供更及时客观全面的信息基础和保障.     

地质统计学变差函数人机对话拟合

变差函数的研究在地质统计学中具有十分重要作用，本文运用界面图形图像处理强的Ｃ￣（＋＋）语言实现了界面友好汉化人机对话变差函数的拟合，主要包括管理菜单的生成，实验变差值的求解，变差图的图形显示，标准函数模型的计算及变差函数人机对话求解等部分。最后对比一下回归分析与人机对话拟合结果。     

Comparing Human and Neural Network Learning of Climate Categories

This study compares how humans and neural networks classify climate types. Human subjects were asked to classify climates from monthly temperature and precipitation patterns. To model their learning process, the same data were used to produce input vectors that trained a pattern associator neural network. Both human subjects and the neural network classified climates accurately after 10 rounds of supervised learning. The neural network successfully modeled the rate of human learning and the ability to learn specific climate categories. Moreover, the neural network weights used to classify climates correspond to distinct visual characteristics in temperature and precipitation. These results suggest that neural networks can model the formation of visual categories.     

An interactive system to assist mineral identification in ore microscopy

An interactive computer system has been developed to assist the mineral identification in ore microscopy. The reference file of the system consists of optical, mechanical, and chemical properties of about 130 ore minerals. The properties are name, chemical formula, color, bireflectance, anisotropism, internal reflection, reflectance at wavelengths of 470, 546, 589, and 650 nm, and polishing hardness and micro-indentation hardness. All the properties except reflectance and microindentation hardness are qualitative or semi-qualitative. Most of the properties are given as characters relative to the more common minerals. This implies that most of the identification processes advance on the basis of the comparison between a subject mineral and coexisting minerals. For this reason, the system asks a user at first to input already identified mineral names. This is quite different from the mineral identification procedures used in petrographic microscopy. To reduce the number of possible minerals, the system presents a series of questions to a user, and the user selects any of the prepared answers according to his observation. The user can also choose any desired question independently of the sequence. The user is expected to be able to recognize some common minerals, such as pyrite, chalcopyrite, galena, and hematite, without the assistance of the system.     

计算机勘探模拟系统的剖面图交互式编辑子系统

计算机的勘探模拟系统在国外已进入实用阶段。目前我们正在做这方面的尝试。在初期设想的计算机勘探系统中很多是在剖面图上进行的,所以剖面图的交互式编辑子系统是整个系统核心的部分,它的主要设计思想是用户通过计算机键盘操作、用类似手工的方法在计算机屏幕上画出一张供自己使用的地质剖面图,并且还有修改、删除、存贮等功能,使用户感到直观、灵活、方便且易于比较。