车载LiDAR点云相连行道树精细分割

针对车载激光雷达获取的行道树树冠点云相连问题,该文提出了基于树木生长模型的距离加权分割方法。首先通过等距原则将相连树粗分割为单株树木,采用求差和拟合算法获得树木高度、胸径等属性信息;再根据树木的理论和实际生长模型,计算得到树冠距离加权值;最后在此基础上对树冠进行迭代处理,实现邻接行道树点云的精确分割。经过不同树种试验验证,该算法能够实现对多棵相连树点云的进一步精确分割。     

利用概率松弛法的城市路网自动匹配

多源空间数据匹配是空间数据集成与互操作,变化检测与数据更新的重要前提。路网数据匹配在导航、智能交通和基于位置服务等领域具有重要的研究意义和实用价值。本文提出一种基于概率松弛方法的城市路网自动匹配方法,该方法首先通过路段间几何差异性估算候选路段的初始概率,然后根据邻接候选匹配路段的兼容性不断更新原概率矩阵直到收敛于某一极小值。最后基于收敛的概率矩阵计算各候选路段的结构相似性,并通过设定相应的规则选取和提炼1: 1, 1: M和M: N匹配对。实验选取中国武汉,瑞士苏黎世地区的OpenStreetMap数据与导航数据进行匹配算法的验证。结果表明：本文算法对非刚性偏差较大的路网数据能达到较高精度,不存在匹配方向性问题,且能够识别1: 0, 1: M和M: N匹配。     

Using the OSPM Model on Pollutant Dispersion in an Urban Street Canyon

An observational campaign was conducted in the street canyon of Zhujiang Road in Nanjing city in 2007.Hourly mean concentrations of PM10 were measured at street and roof levels.The Operational Street Pollution Model(OSPM)street canyon dispersion model was used to calculate the street concentrations and the results were compared with the measurements.The results show that there is good agreement between measured and predicted concentrations.The correlation coecient R2 values(R2 is a measure of the correlation of the predicted and measured time series of concentrations)are 0.5319,0.8044,and 0.6630 for the scatter plots of PM10 corresponding to light wind speed conditions,higher wind speed conditions,and all wind speed conditions,respectively.PM10 concentrations tend to be smaller for the higher wind speed cases and decrease rapidly with increasing wind speed.The presentations of measured and modelled concentration dependence on wind direction show fairly good agreement.PM10 concentrations measured on the windward side are relatively smaller,compared with the corresponding results for the leeward side.This study demonstrates that it is possible to use the OSPM to model PM10 dispersion rules for an urban street canyon.     

Adapting visual methods: action research with Kampala street children

The paper examines the use of four visual 'action' methods for eliciting information from street children about their interactions with the socio-spatial environment. These methods were adapted and used to encourage child-led activities and minimize researcher input. The advantages and constraints associated with the successful implementation of visual methods with children are examined, followed by a brief examination of the 'real' and ethical considerations surrounding their use. The paper concludes that visual methods allow a high level of child-led participation in research, as well as providing a stimulus for eliciting further oral material.     

Street names and the scaling of memory: the politics of commemorating Martin Luther King, Jr within the African American community

Streets named for Martin Luther King, Jr are common yet controversial features in cities across the United States. This paper analyses the politics of naming these streets as a 'scaling of memory'– a socially contested process of determining the geographic extent to which the civil rights leader should be memorialized. Debates over the scaling of King's memory revolve around the size of the named street, the street's level of prominence within a hierarchy of roads, and the degree to which the street transcends the spatial confines of the black community. A street-naming struggle in Eatonton, Georgia (USA) exposes how the scaling of memory can become a point of division and contest within the black community as activists seek to fulfil different political goals. Analysing these intra-racial contests allows for a fuller appreciation of the historical consciousness and geographic agency of African Americans rather than seeing them as a single, monolithic group.     

基于街景影像的城市道路空间舒适度研究

城市道路空间舒适度是综合刻画城市道路环境的量化指标之一,可为城市道路规划建设、城市交通优化、城市环境评估等提供参考.本文以街景影像为基础数据,从空间视角阐述了城市道路空间舒适度内涵,构建了城市道路空间舒适度量化指标,建立了以语义分割和等距采样为基础的城市道路空间舒适度测度方法和空间插值方法,最后以南京市局部城区为样区进...     

分部加权的行道树检测网络构建

基于深度学习目标检测框架,提出了一种端到端的训练网络,用于行道树的自动检测。由于行道树之间的遮挡问题,现有的通用物体检测框架无法直接应用于此任务,为此本文提出了一种树形分部加权模块,以减少严重遮挡造成的错误检测。然后对提出的神经网络进行训练和评估。结果显示,本文所建立的分部加权树木检测网络能够在遮挡条件下,有效地检测出街景图像中的行道树,该方法在各种条件下均具有较高的精度和良好的稳健性。     

街景图片识别城市贫困的适用性——基于广州市中心城区的验证

以广州市为例,选取中心4个区6 670个采样点（涵盖121个社区）的百度街景图片,从城市建成环境特征探讨了城市贫困识别的可能。首先,训练基于深度神经网络的街景图片分类模型后,对街景要素进行语义分割,并通过缓冲区分析统计社区尺度的街景指标;其次,经主成分分析法提取出建筑围合感、植被围合感、天空开阔感和道路开阔感4个街景主因子,并验证其与多重贫困指数（IMD）的相关性;最后,通过采用简单随机抽样法选取61个社区,构建街景预测的多元线性回归模型,对剩余60个社区进行贫困预测,验证街景指标测度城市贫困的度量精度。结果发现,案例社区的多重贫困指数（IMD）与建筑围合感呈正相关,与植被围合感、天空开阔感、道路开阔感呈负相关;从整体看来,街景预测结果与传统城市贫困测度的空间规律基本相符,而且结果通常比传统测度的城市贫困程度高。这是因为受测度内容、社区类型、街道属性等方面的影响,街景识别方式比较适用于判断建成环境较差的贫困社区。街景图片预测有利于刻画城市贫困人群真实的生活环境,便于对城市建成区进行及时监测,在一定程度上可以与传统城市贫困测度相互校正、弥补不足。     

小城镇街道灰尘中氮磷含量研究

以上海市的枫泾镇(FJ)、松江新城(SJ)和朱家角镇(ZJJ)为例,研究了小城镇街道灰尘中氮磷的含量水平和分布特征,结果表明:小城镇街道灰尘中总无机氮的含量为11.9-1040.6μg/g,平均含量为117.5.μg/g;可溶磷的含量为15.8～65.8μg/g,平均含量为48.7μg/g;各小城镇街道灰尘中的氮磷含量表现为:FJ＞ZJJ＞SJ.各小城镇街道灰尘中氮磷含量的空同分布差异较大,位于城镇中心和风景区内采样点的氮磷含量明显较高,而城镇周边乡村地区含量较低.对各小城镇街道灰尘氮磷含量的季节变化分析表明,秋季小城镇街道灰尘中氮磷的含量明显高于冬季和夏季,温度、降雨量可能是其季节变化的主要影响因子.