暴雨内涝影响下的城市道路交通拥挤特征识别

近年来,暴雨内涝频繁发生,常引发严重的城市交通拥堵问题。本文利用自主开发的宏观交通模拟工具,模拟了上海市中心城区50年一遇和100年一遇暴雨强度情景下每条路段的小时交通量,通过计算道路饱和度,研究了不同强度暴雨内涝对中心城区高架出入口和重要道路拥挤程度的影响。结果表明：①100年一遇暴雨内涝对上海市中心城区道路交通服务能力影响显著,可导致7个高架道路出入口关闭,部分出入口严重拥堵;②暴雨内涝对道路拥堵状况影响的差异性明显,变拥挤路段占道路总里程的13.35%,其中一级道路的拥挤程度变化最为明显,如：大连路、武宁路,周家嘴路和长寿路等主要路段服务水平下降。     

Toponym matching through deep neural networks

Toponym matching, i.e. pairing strings that represent the same real-world location, is a fundamental problemfor several practical applications. The current state-of-the-art relies on string similarity metrics, either specifically developed for matching place names or integrated within methods that combine multiple metrics. However, these methods all rely on common sub-strings in order to establish similarity, and they do not effectively capture the character replacements involved in toponym changes due to transliterations or to changes in language and culture over time. In this article, we present a novel matching approach, leveraging a deep neural network to classify pairs of toponyms as either matching or nonmatching. The proposed network architecture uses recurrent nodes to build representations from the sequences of bytes that correspond to the strings that are to be matched. These representations are then combined and passed to feed-forward nodes, finally leading to a classification decision. We present the results of a wide-ranging evaluation on the performance of the proposed method, using a large dataset collected from the GeoNames gazetteer. These results show that the proposed method can significantly outperform individual similarity metrics from previous studies, as well as previous methods based on supervised machine learning for combining multiple metrics.     

Rockfall in the Port Hills of Christchurch: Seismic and non‐seismic fatality risk on roads

Numerous rockfalls released during the 2010–2011 Canterbury earthquake sequence affected vital road sections for local commuters. We quantified rockfall fatality risk on two main routes by adapting a risk approach for roads originally developed for snow avalanche risk. We present results of the collective and individual fatality risks for traffic flow and waiting traffic. Waiting traffic scenarios particularly address the critical spatial‐temporal dynamics of risk, which should be acknowledged in operational risk management. Comparing our results with other risks commonly experienced in New Zealand indicates that local rockfall risk is close to tolerability thresholds and likely exceeds acceptable risk.     

基于全球生产网络视角下的零售供应链升级与转型——以鲜奶和大豆油为例

在过去的10年中,快速的零售业全球化进程已经在经济地理学研究中引起了强烈的关注。依托全球生产网络分析框架,通过阐述权利与价值在两个食品供应链（鲜奶和食用油）中发生的迁移与更迭,揭示出跨国零售企业进入中国市场前后,本土供应链发生的转型与升级。同时,就供应链转型所具有的五个普适性特征,包括集中化采购、专业批发商的涌现、物流与信息系统升级、执行企业标准和发展自有品牌,在选取的案例产品中进行分析与比较,以此来解释供应链重组过程中的影响因素。结果表明：应采用动态的视角来研究零售供应链中的权利与价值,以此才能更好地揭示新兴市场中供应链网络转型升级的复杂性和不均衡性。     

路域植被等效水厚度估算模型研究

植被等效水厚度对路域生态环境的监测评估具有重要意义。本研究以湖南醴潭高速一段为研究对象,以地面实测光谱和等效水厚度以及PRO4SAIL模拟光谱和模拟等效水厚度为数据源,利用PRO4SAIL冠层模型模拟光谱与地面实测光谱建立12种常用水分指数,引入随机森林算法对水分指数与等效水厚度进行重要性分析,得到12种水分指数的重要性排序;利用调整R ²确定建立等效水厚度估算模型中输入水分指数的最佳个数;在优选水分指数基础上,以PRO4SAIL模拟光谱计算得到水分指数和等效水厚度为训练集,分别构建随机森林耦合偏最小二乘（RF-PLS）、随机森林耦合支持向量机（RF-SVM）和随机森林耦合遗传算法优化支持向量机（RF-GA-SVM）等效水估算模型,并用地面实测等效水厚度对估算模型进行精度验证与分析。结果表明：RF-SVM估算模型中输入重要性前9的水分指数（NDWI、NMDI、SRWI、SR、NDII、WI、DWI、MSI、SAVI）时,调整R ²最高,验证集决定系数为0.8877;RF-PLS和RF-GA-SVM估算模型中输入重要性前4的水分指数（NDWI、NMDI、SRWI、SR）时,调整R ²最高,验证集决定系数分别为0.8053、0.8952,其中RF-GA-SVM模型估算等效水厚度效果最佳,其精度满足路域植被等效水厚度监测要求。本文研究成果为等效水厚度估算提供一种有效且精确的方法,同时为发展基于高光谱遥感的路域环境监测提供重要支撑。     

基于Anchor-free的交通标志检测

交通标志检测是自动驾驶中的重要研究方向,实时准确地从街景图像中检测交通标志对实现自动驾驶及智慧城市的发展具有重要意义。传统的算法基于颜色、形状特征进行检测,只能提取特定种类的交通标志,算法无法同时检测不同类型的交通标志。基于图像特征+机器学习分类器的算法需要人工设计特征,算法速度较慢。主流的基于深度学习的方法多基于先验框,在网络设计上引入了额外的超参数,且在训练过程中产生过量的冗余边界框,容易造成正负样本不平衡。本文受Anchor-free思想的启发,引用YOLO检测器直接回归物体边界框的思路,提出一种基于Anchor-free的实时交通标志检测网络AF-TSD（Anchor-free Traffic Sign Detection）。AF-TSD摒弃了先验框的设计,并引入自适应采样位置可变卷积与注意力机制,大大提高网络的特征表达能力。本文开展大量对比实验,实验结果表明本文提出的AF-TSD交通标志检测网络速度接近主流算法,但精度优于主流算法,在德国GTSDB交通标志检测数据集上取得了96.80%的精度,检测速度平均单张图片32 ms,达到实时检测的要求。     

改进型算法的高分辨率遥感影像道路灾害信息提取

对灾后道路受灾信息获取,分析其交通通行能力,是救灾关键之一。本文针对传统高分辨率遥感影像道路信息提取方法效率低的问题,在面向对象算法基础上,提出了一种改进型的矢量与影像叠加的算法,对灾害道路进行信息提取。试验结果表明,该算法提高了灾害道路信息提取的效率与精度。     

大规模GNSS网分布式存储与解算方法

针对GNSS大网在海量数据存储和解算方面面临的挑战,设计了从基础设施、数据管理、计算和服务到应用的GNSS分布式存储与解算体系结构;基于HDFS改进了GNSS数据存储目录结构;基于HBase设计了GNSS数据存储表结构;基于MapReduce提出了GNSS大网分区分布式解算策略。实验搭建了分布式平台,实现了GNSS数据分布式存储、并行检索和发布以及GNSS大网分区分布式解算,存储和解算性能均得到了较大提高。结果表明,所提出的方法可用于GNSS大网数据存储和解算。     

Monitoring permeable paver runoff with an open-innovation geospatial sensor network

Sensor networks are an essential tool for environmental scientists. As scientists and engineers are beginning to utilize these new methods and devices in their fieldwork, they need to be actively involved in the future of sensor-networking development. Continued sensor network innovation is important for improved standardization, affordability, and interoperability. This article uses a storm water case study to outline an end-to-end open-innovation sensor network. Open innovation by scientists, engineers, and entities is the collaborative process of creating value for this project in permeable paver runoff data and advances within sensor networking. This article focuses on the technical implementation of the near–real-time location and temporally aware sensor network. Data are streamed in near–real-time with subliter precision to the cloud using common off-the-shelf routers. The sensors use Maxim's 1-wire? protocol, and the unique digital serial numbers confirm the data. The data retrieved compare residence times within the permeable paver catchment basins and the control basin. Sensor network advances are made by bridging the gap between sensor protocols and communication systems. These advances enable the development of open-source representational state transfer web services. Our successful implementation serves as an example for others to study and expand upon for a variety of monitoring solutions.     

Alteration-weakening leading to localized deformation in a damage aureole adjacent to a dormant shear zone

Deformation adjacent to faults and shear zones is traditionally thought to correlate with slip. Inherited structures may control damage geometry, localizing fluid flow and deformation in a damage aureole around structures, even after displacement has ceased. In this paper we document a post-shearing anastomosing foliation and fracture network that developed to one side of the Mesoarchean Marmion Shear Zone. This fracture network hosts the low-grade, disseminated Hammond Reef gold deposit. The shear zone juxtaposed a greenstone belt against tonalite gneiss and was locked by an intrusion that was emplaced during the final stages of suturing. After cessation of activity, fluids channeled along fault- and intrusion-related fractures led to the pervasive sericitization of feldspars. Foliated zones resulted from flattening in the weaker sericite-rich tonalite during progressive alteration without any change in the regional NW-SE shortening direction. The anastomosing pattern may have been inherited from an earlier ductile fabric, but sericite alteration and flattening fabrics all formed post-shearing. Thus, the apparent foliated fracture network adjacent to the Marmion Shear Zone is a second-order effect of shear-related damage, distinct in time from shear activity, adjacent to an effectively dormant shear zone. This phenomenon has implications for understanding the relative timing of fault zone activity, alteration and (in this case) gold mineralization related to long-term fault zone permeability.