单变量特征选择的苏北地区主要农作物遥感识别

遥感识别多源特征综合和特征优选是提高遥感影像分类精度的关键技术。农作物遥感识别中,识别特征的相对单一和数量过多均会导致作物识别精度不理想。随机森林(random forests)采用分类与回归树(CART)算法来生成分类树,结合了bagging和随机选择特征变量的优点,是一种有效的分类方法。单变量特征选择(univariate feature selection)能够对每一个待分类的特征进行测试,衡量该特征和响应变量之间的关系,根据得分舍弃不好的特征,优选得到的特征用于分类。本文基于随机森林和单变量特征选择,利用多时相光谱信息、植被指数信息、纹理信息及波段差值信息,设计多组分类实验方案,对江苏省泗洪县的高分一号(GF-1)和环境一号(HJ-1A)影像进行分类研究,旨在选择最佳的分类方案对实验区主要农作物进行识别和提取。实验结果表明:(1)多源信息综合的农作物分类精度明显高于单一的原始光谱特征分类,说明不同类型特征的引入能改善分类效果;(2)基于单变量特征选择算法的优选特征分类效果最佳,总体精度97.07%,Kappa系数0.96,表明了特征优选在降低维度的同时,也保证了较高的分类精度。随机森林和单变量特征选择结合的方法可以提高遥感影像的分类精度,为农作物的识别和提取研究提供了有效的方法。     

多目标遗传算法改进的加权多点灰色模型

针对传统多点灰色预测模型MGM(1,n)白化背景值构造方法不合理性导致模型往往不符合变形体实际情况的问题,该文提出了一种基于遗传算法的加权MGM(1,n)模型。引入白化背景值最佳生成权值矩阵替换传统模型背景值构造公式中的紧邻均值生成权阵,较好地顾及变形区域内多监测点变形趋势的突变性与不规则性,弥补了线性系统MGM(1,n)模型在非线性动力学系统变形预测分析应用中的不足;建立多目标优化实数编码遗传算法,实现背景值最优构造权阵的迭代搜索。基于仿真和工程实例数据的建模结果表明:改进模型较传统MGM(1,n)模型预测精度提高,抗噪声干扰能力增强。     

改进灰色人工神经网络模型的超高层建筑变形预测

针对灰色人工神经网络模型初始化权值和阈值的随机性导致易产生误差积累和过拟合的缺陷,该文利用遗传算法的全局优化能力训练灰色人工神经网络模型的权值和阈值,构建了基于遗传算法的灰色人工神经网络超高层建筑物变形预测模型。结合长沙北辰新河A1超高层建筑变形监测实例,用该文所提模型与灰色人工神经网络模型分别进行变形数据的处理分析和预测。实验结果表明,该文提出的模型具有更好的预测精度,预测趋势也更加逼近实际测量结果。     

时空理论变异函数模型及其精度影响

针对已有的时空模型研究深度不够的问题,该文主要研究两类时空理论变异函数模型,即分离和非分离模型,并分析其对时空预测精度的影响。首先,阐述了两类模型中较为流行的6种时空理论子模型及其参数物理意义;最后以2014年山东省PM2.5日均浓度为例,给出各模型理论表达式,并在Matlab中对各模型运用遗传算法进行参数拟合和时空预测精度验证。实验结果显示,6种模型的整体预测精度顺序与模型拟合精度顺序相当,说明时空理论模型同空间理论模型一样,其模型拟合精度与后续预测精度成正相关。     

A hybrid genetic algorithm with local optimiser improves calibration of a vegetation change cellular automata model

Cellular automata (CA) models are commonly used to model vegetation dynamics, with the genetic algorithm (GA) being one method of calibration. This article investigates different GA settings, as well as the combination of a GA with a local optimiser to improve the calibration effort. The case study is a pattern-calibrated CA to model vegetation regrowth in central Victoria, Australia. We tested 16 GA models, varying population size, mutation rate, and level of allowable mutation. We also investigated the effect of applying a local optimiser, the Nelder?Mead Downhill Simplex (NMDS) at GA convergence. We found that using a decreasing mutation rate can reduce computational cost while avoiding premature GA convergence, while increasing population size does not make the GA more efficient. The hybrid GA-NMDS can also reduce computational cost compared to a GA alone, while also improving the calibration metric. We conclude that careful consideration of GA settings, including population size and mutation rate, and in particular the addition of a local optimiser, can positively impact the efficiency and success of the GA algorithm, which can in turn lead to improved simulations using a well-calibrated CA model.     

Modeling urban expansion by using variable weights logistic cellular automata: a case study of Nanjing,China

Simulation models based on cellular automata (CA) are widely used for understanding and simulating complex urban expansion process. Among these models, logistic CA (LCA) is commonly adopted. However, the performance of LCA models is often limited because the fixed coefficients obtained from binary logistic regression do not reflect the spatiotemporal heterogeneity of transition rules. Therefore, we propose a variable weights LCA (VW-LCA) model with dynamic transition rules. The regression coefficients in this VW-LCA model are based on VW by incorporating a genetic algorithm in a conventional LCA. The VW-LCA model and the conventional LCA model were both used to simulate urban expansion in Nanjing, China. The models were calibrated with data for the period 2000–2007 and validated for the period 2007–2013. The results showed that the VW-LCA model performed better than the LCA model in terms of both visual inspection and key indicators. For example, kappa, accuracy of urban land and figure of merit for the simulation results of 2013 increased by 3.26%, 2.96% and 4.44%, respectively. The VW-LCA model performs relatively better compared with other improved LCA models that are suggested in literature.     

NDVI、NDMI与地表温度关系的对比研究

通过研究归一化植被指数(NDVI)、归一化水汽指数(NDMI)与地表温度(LST)的相关关系,对比NDVI和NDMI定量分析LST的适宜程度。以Landsat 8遥感影像为数据源,以郑汴都市区为例,反演LST,计算NDVI和NDMI。从整体、分区、像元等不同层面分析NDVI和NDMI与LST的相关关系,并利用GEO-Da软件,分析150 m、300 m、450 m三种不同采样间隔数据的NDVI、NDMI与LST的空间相关性。主要结论为：①NDVI与LST的线性拟合度较差,而NDMI与LST具有很强的线性关系,剖面分析显示NDMI与LST呈显著负相关关系。②缓冲区分析结果表明,随着距城市中心距离的增加,用地类型增多,LST和NDMI之间的相关性逐渐增强。③在Moran's I空间相关性分析中,不同采样间隔下两指数与LST的负相关关系均比较明显,但由于水体在2个指数中数值的差异,使NDMI与LST表现出更强的空间负相关性,而NDVI与LST空间相关性则相对较弱。因此,总的来说NDMI是热环境研究的有效指标,与NDVI相比,NDMI与LST的负相关关系更强、更稳定,也更适宜于对LST的定量分析。     

二维地质体的瞬变电磁场响应特征

直接从时间域出发 ,应用时域有限差分方法 (FDTD)分析地下和地面的瞬变响应。由于二维情况便于图形表示 ,采取了线源二维地电模型。通过对二层均匀地电结构和含二维低阻体时地面垂直感生电动势曲线的计算 ,说明瞬变场响应有滞后现象 ,即异常地质体一旦引起瞬变响应 ,就会往后延时 ,并使响应延续较长时间。在实际的工程勘察中 ,曾用 3～ 5ms以后的延迟时间探测过较浅的异常体 ,如 2 0m至 150m深度范围内的地下老窑采空区 ,并取得了较好的探测效果 ,上述数字模拟结果为以往的实践提供了理论依据     

An Improved Method for Calculating Paleoheat Flow from Vitrinite Reflectance Profiles

lNTRODUCTIONIt is wel1 established that the vitrinite reflectance (R<, ) de-pends on the temperature history that a basin has undergone.In general, Ro is a function of maximum paleotemperature andeffective heating time. Thus, the paleotemperature of sedi-ments and the value of vitrinite ref1ectance are controlled by theflow of heat Q(t) into the basin and by the thermal parametersof the sediments. Therefore, in principle, it is possible to cal-culate the thermal vitrinite reflectance of …     

乌拉山岩金矿田微斜长石成因成矿信息研究及其意义

乌拉山含金矿脉赋存于新太古代乌拉山群高级变质岩区（变质程度以角闪岩为主，可达麻粒岩相），受控于乌拉山断裂构造带。区内出露了大桦背花岗岩体以及大量不同成因的长英质体，含金矿脉主要由中等一最大微斜长石，石英，黄铁矿和微晶自然金等矿物组成。对矿脉微斜长石有关信息作了Ｉ值信息量，对应－聚类及成因判别的分析，研究表明成矿作用可分为两个阶段，矿脉系岩浆热液成因。矿脉微斜长石钾长石组分的摩尔分数，有序度（Ｚ）和