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％。     

Methodological problems of intercultural comparisons of retail environments

The internationalising activities of retailers have tended to precede the theoretical contributions of academics. This is particularly the case for the cultural context within which retail internationalisation takes place. Following an examination of the growth of international retailing activity, the article reviews generic difficulties in studying inter-cultural differences before considering the specific steps which retail practitioners have taken to understand the comparative characteristics of retail and consumer markets through the growth of geodemographic and lifestyle analysis. Finally, we relate one of the most significant integrative intercultural theoretical instruments to the case of retail internationalisation within Western Europe. We find that significant insights into retailer behaviour and strategy can be gained from an understanding of the culturally-specific characteristics of the activity. This revised version was published online in July 2006 with corrections to the Cover Date.     

江苏下扬子区中—古生界的成藏条件及勘探思路

江苏下扬子区的海相中－古生界构造演化经历了前扬子板块、稳定地台和西太平洋活动陆缘等三大阶段。油气地质条件具有多套烃源岩、两类主要储层、四套区域盖层、多种成藏类型和多期成烃等特征。原始生烃量以下古生界为主,下寒武统和下志留统的生烃量约占总生烃量的６０％￣８０％。该领域的勘探应本着“整体评价、区域展开、重点突破”的原则,做到“六个结合”和“五个转变”,加快勘探节奏,重在发现,坚持以下古生界为主,大中型     

旅游地品牌特征塑造初探--以香格里拉旅游品牌为例

香格里拉是一个公共品牌，是系列旅游产品的组合，香格里拉品牌具有鲜明的特征，其特征通过品牌的个性和定位体现，品牌的个性是通过文化性和奇特性表现，信念性是品牌的定位，具体分析了香格里拉品牌的资源本体的脆弱性，旅游产品开发层次较低，品牌崇信度不确定性和品牌定位不准等问题，最后对香格里拉品牌特征的塑造，提出了建议；树立品牌意识，加强品牌管理；加强品牌定位－突出个性，构筑品牌的保护圈，实现品牌特征的接触点。     

A comparative study of extensive machine learning models for predicting long-term monthly rainfall with an ensemble of climatic and meteorological predictors

Rainfall prediction is of vital importance in water resources management. Accurate long-term rainfall prediction remains an open and challenging problem. Machine learning techniques, as an increasingly popular approach, provide an attractive alternative to traditional methods. The main objective of this study was to improve the prediction accuracy of machine learning-based methods for monthly rainfall, and to improve the understanding of the role of large-scale climatic variables and local meteorological variables in rainfall prediction. One regression model autoregressive integrated moving average model (ARIMA) and five state-of-the-art machine learning algorithms, including artificial neural networks, support vector machine, random forest (RF), gradient boosting regression, and dual-stage attention-based recurrent neural network, were implemented for monthly rainfall prediction over 25 stations in the East China region. The results showed that the ML models outperformed ARIMA model, and RF relatively outperformed other models. Local meteorological variables, humidity, and sunshine duration, were the most important predictors in improving prediction accuracy. 4-month lagged Western North Pacific Monsoon had higher importance than other large-scale climatic variables. The overall output of rainfall prediction was scalable and could be readily generalized to other regions.     

成人英语学习动机行为调查与教学对策研究

动机是激励人们行动的心理动因.英语学习动机是推动英语学习的内部动因,成人英语学习动机是复杂的.了解他们的英语学习动机特点有利于教师改进教学手段和策略,激发与培养其学习的主动性和积极性,发挥其学习潜能,从而提高英语学习的效果.运用定量研究的方法,在年轻干部学员中进行学习动机调查,分析了其学习动机的特点,并提出成人英语教学的对策.     

基于高清工业相机的盾构隧道裂缝智能识别算法分析

裂缝一直是隧道病害的重点检测对象,但传统人工巡检仅能通过肉眼发现后记录,人工识别精准度与效率完全取决于个人经验判断,无信息化手段辅助,作业效率识别精度亟待提升。针对以上问题,本文借助高清工业相机成像分辨率高、采集速度快等特点,将高清工业相机部署于轨道车上获取隧道表面裂缝病害信息,大幅提高了隧道裂缝识别效率,将识别精度提升至0.2 mm,同时融入优化的Cascade R-CNN算法,在有监督情况下训练隧道裂缝样本,最终实现了隧道裂缝病害的高效提取,同时研发了一套包含硬件数据采集、数据处理软件、数据管理平台的裂缝病害识别路线,真正意义上破除了识别慢、精度低、靠经验、难管理的技术壁垒。     

顾及几何特征和拓扑连续性的道路网匹配方法

已有的道路网匹配方法需要一定的人为干预和控制,仍无法实现匹配自动化,而且对于具有一定尺度差异的数据而言,受不相关道路对象的影响,匹配结果中会存在较多的误匹配。为此,本文提出一种顾及几何特征和拓扑连续性的由粗匹配到精匹配的分层匹配策略。粗匹配阶段,利用短边中位数Hausdorff距离计算匹配对象的相似度,再以道路网弧段端点的近邻分析结果作为匹配阈值,判断匹配对象是否为同名实体;精匹配阶段,将粗匹配阶段评价结果中离散的同名实体构建为完整的道路Stroke,即以追踪的方式剔除误匹配和添加漏匹配。利用宿城区不同尺度的道路网对所提算法进行验证,结果表明,该算法具有较好的匹配效果,能够实现匹配自动化。     

建筑物立面分类与DIM点云滤波

针对目前基于近景摄影测量方法构建建筑物立面模型过程中因密集影像匹配（DIM）点云噪声所引起的建筑物立面TIN网格模型畸变问题,本文借鉴机器学习中样本学习的思想,对建筑物立面进行了分类并对DIM点云提出了相应的滤波方法,以达到去除DIM点云噪声和改善其TIN网格模型畸变的目的。其中,针对平面结构立面,采取先对点云样本进行学习计算构建数学立面模型所需参数,再对该立面模型设定阈值并对其点云进行滤波处理的方法;针对曲面结构立面,则结合DIM点云特性先将点云样本分类标记归为立面点与非立面点,再进行样本特征值学习,使用Logistic回归算法迭代计算求解最佳回归系数,从而构建滤波分类器的方法对立面点云进行滤波处理。试验结果表明,本文滤波处理方法能将立面DIM点云噪声有效识别并去除,而且使用该方法处理后所得点云构建的建筑物立面TIN网格模型精细化程度得到有效提高,模型质量得到明显改善。     

深度学习PaddlePaddle框架支持下的遥感智能视觉平台研究与实现

百度深度学习PaddlePaddle框架支持下的遥感智能视觉平台,能够运用深度学习技术实现遥感影像的智能建模、训练和解译。本文通过深入分析PaddlePaddle图像分割模型库PaddleSeg的图像处理深度学习算法模型DeepLabV3+、U²-Net及RetinaNet,开发设计了遥感智能视觉平台,实现了遥感影像的地块分割、变化检测和斜框检测等专业功能。研究表明：遥感智能视觉平台提取的图斑总面积是目视解译的80%、有效图斑比例为76%、错误图斑比例为18%,实现了快速有效的遥感图像智能处理。