基于贝叶斯逻辑回归模型研究百度街景图像微观建成环境因素对街面犯罪的影响

街面犯罪对公众的生活安全构成一定的威胁。以往对于公共盗窃和寻衅滋事等街面犯罪的研究往往停留在社区甚至更宏观的层面,难以向微观尺度深入,它们忽略了通过环境设计预防犯罪(Crime Prevention Through Environmental Design,CPTED)理论中所主张的地址级的建成环境的精确特征。地址级的微观建成环境被广泛认为对各类犯罪的发生有着直接或间接的影响,然而对微观建成环境的度量一直是一个挑战。先前的大多数研究都是通过调查样本来表征建成环境,会受到两方面的限制:(1)建成环境特征描述不完整的限制;(2)数据在空间覆盖方面具有稀疏性的限制。百度街景图像作为一个新的数据来源,可以被用来提取地址级的微型环境的建成特征,从而使犯罪研究可以聚焦在更微观的尺度中。本研究使用深度学习全卷积图像分割算法从百度街景图像中提取地理位置的环境变量,共选取树木、通车道路、人行道等8种变量来表现研究区微观建成环境的差异。在控制了与街面犯罪有关的其他因素后,采用贝叶斯逻辑回归模型来评估微观建成环境影响因素对公共盗窃和寻衅滋事案件的影响。结果表明加入了微观建成环境物理特征之后的模型表现更好。对...     

GIS的不同次区域之间相似性度量

 GIS属性相似度量有相似系数及以空间位置的空间自相关。这2种方法都不能计算全域为背景的次区域相似性问题,而以隶属函数的粗糙度量很好地解决了这个问题。本文利用粗糙集计算了GIS的离散型数据的次区域相似度量。利用邻域粗糙集计算了连续属性值的上下近似逼近,并利用邻域信息粒子计算连续属性值的粗糙隶属问题,从而计算连续属性值次区域相似度量问题。本文方法能度量次区域而不是全域或单个元素之间的空间相关与自相关问题,考虑了全区域背景下的GIS相似问题。并针对GIS连续属性值,利用距离函数和邻域粗糙集来划分连续属性的上下近似,以及分类问题,提出一种基于邻域信息粒子的粗糙隶属函数。最后利用粗糙相似度量公式度量GIS次区域的相似问题。     

Spatial Differentiation and Influencing Mechanism of Medical Care Accessibility in Beijing: A Migrant Equality Perspective

Spatial equality of access to basic public services, especially medical care services which are directly related to life safety, is the first step to achieve the goal of equalization of basic public services for all the people proposed by central government of China. Using the spatial analysis and the statistical analysis, this study evaluates the spatial differentiation of medical care facilities accessibility by constructing STT (Shortest Travel-Time) and SAI (Spatial Accessibility Index). And then this study explores the neighborhood effects on the medical care facilities accessibility in Beijing, with a particular focus on the effect of neighborhood migrant proportion by constructing spatial dependent regression model. The spatial accessibility analysis of medical care facilities show that the spatial distribution of medical care facilities was basically consistent with administrative regions but not with population demands. Bivariate LISA cluster maps identify that suburban areas are the overlapped clusters of high percent of migrants and limited medical care services. This is associated with the public service allocation rule in China, which stresses equality within urban areas and within rural areas but overlooks equality between urban areas and rural areas; and stresses local resident demands but overlooks migrant demands. To estimate the effects on medical care accessibility of neighborhood migrant proportion, spatial dependence models are applied due to spatial dependence of accessibility of medical care facilities. The regression results show that neighborhoods with high percent of migrants, even conditioning on neighborhood SES, are related to limited spatial accessibility of medical care services. Besides neighborhood characteristics, another important factor influencing spatial accessibility of medical care services is the process of spatial spillover effects. This indicates that the attenuate accessibility of medical care services for migrants is not only because of their own constraints but also because of their proximity to other disadvantaged neighborhoods. Therefore, it is urgently needed to increase the medical facilities in the suburban areas, to take into account migrants’ demands and to reduce residential segregation between local residents and migrants for local governments to achieve the goal of equalization of medical care service.     

典型贫困山区旅游景点可达性评价方法研究

旅游景点的可达性水平不仅可以反映游客到达的方便程度,也用来衡量区域旅游业发展的潜力,而对于贫困山区旅游景点的可达性研究还可为旅游扶贫的开展工作提供理论参考。本文聚焦典型贫困山区,探究交通成本及复杂地形特征对旅游景点可达性的影响。首先,文章通过连接度指数和通达度指数的相关分析,了解研究区旅游网络的整体连通状况;然后,运用累积耗费距离法,分别基于不同等级交通因素和坡度、海拔等地形因素,对研究区旅游景点可达性和县级行政单元整体可达性进行对比分析,并提出了一种综合考虑交通因素和地形因素的可达性评价方法。研究表明：交通因素分析法弱化了水域对于可达性的阻隔作用;地形因素分析法忽视了交通条件对于可达性的影响;而综合考虑不同等级交通因素和复杂地形因素的可达性评价方法,既可有效识别水域对于贫困山区旅游景点可达性的阻隔作用,也体现了交通条件的改善对于自然环境下可达性水平的提升作用。     

基于改进引力模型的广场公园可达性研究

城市化的迅速发展使越来越多的自然环境被城市建筑所代替,随着居民生活水平的逐渐提升,居民开始越来越关注广场公园的休闲游憩功能,广场公园的可达性逐渐成为现代人选择居住小区的重要指标之一,如何客观真实地评价广场公园的可达性是学者们长期关注的问题之一。本文分析了距离摩擦系数β从0.6~2.2以步长为0.2变化时对可达性结果的影响,同时为了准确测定广场公园的可达性,将旅游域模型的思想首次应用至可达性研究中,对引力模型中的距离衰减系数β进行估计,最后以武汉市中心城区为例进行验证。结果表明：① 距离摩擦系数β较小的变化将对可达性结果产生较大的影响,随着β的增加可达性计算值出现最大值越来越大、最小值越来越小、标准差越来越大的现象;② 基于旅游域模型的思想构建参数估计方法并选取5组不同实验对引力模型中的距离摩擦系数β进行估算,最终通过求取不同实验组的均值得β=1.70时的可达性评价结果更加可靠;③ 通过武汉市中心城区广场公园可达性的实例验证,结果表明优化后的模型在考虑广场公园规模与人口规模影响因子的同时,可以充分体现广场公园吸引力在距离衰减上的变化,更加客观真实地评价广场公园的可达性。     

Land use intensity dynamics in the Andhikhola watershed,middle hill of Nepal

Land use intensity is a valuable concept to understand integrated land use system, which is unlike the traditional approach of analysis that often examines one or a few aspects of land use disregarding multidimensionality of the intensification process in the complex land system. Land use intensity is based on an integrative conceptual framework focusing on both inputs to and outputs from the land. Geographers' non-stationary data-analysis technique is very suitable for most of the spatial data analysis. Our study was carried out in the northeast part of the Andhikhola watershed lying in the Middle Hills of Nepal, where over the last two decades, heavy loss of labor due to outmigration of rural farmers and increasing urbanization in the relatively easy accessible lowland areas has caused agricultural land abandonment. Our intention in this study was to ascertain factors of spatial pattern of intensity dynamism between human and nature relationships in the integrated traditional agricultural system. High resolution aerial photo and multispectral satellite image were used to derive data on land use and land cover. In addition, field verification, information collected from the field and census report were other data sources. Explanatory variables were derived from those digital and analogue data. Ordinary Least Square(OLS) technique was used for filtering of the variables. Geographically Weighted Regression(GWR) model was used to identify major determining factors of land use intensity dynamics. Moran's I technique was used for model validation. GWR model was executed to identify the strength of explanatory variables explaining change of land use intensity. Accordingly, 10 variables were identified having the greatest strength to explain land use intensity change in the study area, of which physical variables such as slope gradient, temperature and solar radiation revealed the highest strength followed by variables of accessibility and natural resource. Depopulation in recent decades has been a major driver of land use intensity change but spatial variability of land use intensity was highly controlled by physical suitability, accessibility and availability of natural resources.     

不同测算方法下行程时间不确定性对可达性的影响分析

行程时间不确定性导致了可达性随时间的变化,相关研究表明忽略行程时间不确定性会高估可达性水平。既有可达性研究往往用行程时间可靠性表示行程时间不确定性,但未考虑不同可达性模型结果的差异以及行程时间可靠性价值。本文结合各OD之间的行程时间分布特征,构建方差型的行程时间可靠性来描述行程时间不确定性,并进一步将行程时间可靠性纳入到广义出行时间成本中,建立了时间距离模型、潜力模型、累计机会模型和高斯模型4种基于位置的可达性测算方法,以比较在不同测算方法下,行程时间不确定性对可达性的影响。深圳的案例研究表明：(1)忽略行程时间不确定性会使全区域的可达性至少被高估5.04%,最大被高估95.04%。潜力模型、时间距离模型、累计机会模型和高斯模型的高估幅度由低到高;(2)行程时间不确定性对可达性的影响存在阈值效应,阈值越高,可达性受影响的程度越小;(3)从空间分布来看,行程时间不确定性对可达性水平高和低的区域都有一定影响。若不考虑行程时间不确定性,可达性高的区域高估值大,而在可达性低的区域,可达性高估的百分比较大,高估百分比中位数的差异程度最大可达77.1%;(4)行程时间不确定性对潜力模型可达性分类的...     

等时线模型支持下的深圳市综合医院空间可达性测度分析

可达性是在空间分布格局、设施约束等综合条件下的空间决策问题,等时线模型为该决策分析可提供方法支持,它是以时间地理为框架,用时间成本来衡量时空可达性的重要模型。面向兴趣点（Point of Interest, POI）可达性分析通常包括2种形式：① 基于“供给”角度从特定POI出发的服务可达分析;② 基于“需求”角度从任意空间位置出发获得最近服务的便利度分析。目前的研究从“供给”角度探求特定POI的服务范围居多,对“需求”角度的等时线分析关注不够。本研究针对这一缺陷,开展面向服务需求的兴趣点可达性等时线分析,考察任意位置达到某类服务设施最近点的时间代价。在几何度量方法上,本研究利用了网络空间Voronoi图模型,与传统欧氏空间Voronoi图相比,本方法顾及了街道网络对空间通达的直接影响。本研究发现了等时线的形态、坡度等特征能够反映POI的疏密分布和道路网的通达性。     

基于主成分Logistic模型的商圈吸引力因子分析

重力模型是地理学的经典模型之一,常用于解释和预测人口、货物、交通、财富等的移动现象。其中Huff模型被广泛用于预测商圈规模,但目前多数研究以商场的面积作为吸引力的度量指标,不能准确反映商场真实的吸引力,造成模型预测的误差。本文针对Huff模型常使用规模表达商圈吸引力的不足,在对购物中心吸引力进行总结的基础上,选择了深圳市5个具有代表性的大型购物中心（或商圈）进行问卷调查。利用因子分析方法识别出6个影响购物中心吸引力的因子：可达性因子、环境因子、私家车出行因子、多功能因子、商品因子和个人特征因子。基于主成分Logistic模型,分析购物中心吸引力的显著影响因素及权重。最后,通过对比不同参照系下购物中心的显著性因素,对商圈吸引力因子进行可靠性检验。本文的分析方法及结果可为Huff模型等的精细化建模提供基础。     

Spatial evolution and locational determinants of high-tech industries in Beijing

Using datasets on high-tech industries in Beijing as empirical studies, this paper attempts to interpret spatial shift of high-tech manufacturing firms and to examine the main determinants that have had the greatest effect on this spatial evolution. We aimed at merging these two aspects by using firm level databases in 1996 and 2010. To explain spatial change of the high-tech firms in Beijing, the Kernel density estimation method was used for hotspot analysis and detection by comparing their locations in 1996 and 2010, through which spatial features and their temporal changes could be approximately plotted. Furthermore, to provide quantitative results, Ripley′s K-function was used as an instrument to reveal spatial shift and the dispersion distance of high-tech manufacturing firms in Beijing. By employing a negative binominal regression model, we evaluated the main determinants that have significantly affected the spatial evolution of high-tech manufacturing firms and compared differential influence of these locational factors on overall high-tech firms and each sub-sectors. The empirical analysis shows that high-tech industries in Beijing, in general, have evident agglomeration characteristics, and that the hotspot has shifted from the central city to suburban areas. In combination with the Ripley index, this study concludes that high-tech firms are now more scattered in metropolitan areas of Beijing as compared with 1996. The results of regression model indicate that the firms′ locational decisions are significantly influenced by the spatial planning and regulation policies of the municipal government. In addition, market processes involving transportation accessibility and agglomeration economy have been found to be important in explaining the dynamics of locational variation of high-tech manufacturing firms in Beijing. Research into how markets and the government interact to determine the location of high-tech manufacturing production will be helpful for policymakers to enact effective policies toward a more efficient urban spatial structure.     

基于空间场-水文模型的城市影响腹地界定及其空间演变分析——以河南省地级城市为例

城市影响腹地范围的划分,在区域与城市规划中有着重要的理论和实际意义。本文从"城市结节性"和"空间交通可达性"两方面改进场强模型,采用主成分分析法与指标体系计算河南省城市结节性指数,利用累积耗费距离测算空间可达性,借助k阶数据场模型与水文分析模型,综合测度了1991年和2010年河南省17个地级城市的影响腹地范围及空间演变特征。研究表明,20年间河南省地级城市的平均可达本文时间从1991年的45.41 min缩短为2010年的33.03 min,空间场能显著增长且空间分异性显著。南阳的腹地面积增加最大,信阳的腹地面积缩小最大;安阳腹地面积增加率和漯河腹地面积减少率最大。郑州的腹地范围与其行政辖区偏移度最大。     

Equity of Accessibility to Health Care Services and Identification of Underserved Areas

Characterizing the availability of opportunities to residents has been a long-term aim in health care geographic investigation.It is important to measure the degree of inequity in health care accessibility and to identify underserved areas, due to the uneven distribution of health care services. In this study, JavaScript was used to calculate travel time based on Amap, as this can provide a more reliable data support to measure the health care accessibility in Xi’an communities, China. Based on the overall equity, herein, an attempt was made to quantify the equity of health care accessibility, and to identify health care underserved areas inside the different communities. Results show that the accessibility to low-level health care services is high in the northern areas and low in the southern areas, while the accessibility to high-level and comprehensive health care services shows a clear core-periphery spatial structure. Moreover, the overall equity of the health care accessibility is relatively low, and the inequity of high-level health care accessibility is further aggravated.Furthermore, the quantified equity of accessibility to high-level and comprehensive health care services in the central urban areas is better;however low-level health care services are relatively inadequate. There are significant differences among health care underserved areas, in particular, for the worst equity and the lowest accessibility areas(A1) and the worse equity and the lowest accessibility areas(B1) in high-level underserved areas. Notably, the sharing of health care services and the reasonable flow of health technical personnel among different levels of health institutions can make the high-level health care services in the central urban areas have a greater trickle effect on the surrounding areas.     

Evaluating mountain water scarcity on the county scale: a case study of Dongchuan District,Kunming, China

Mountain water, which contributes 50% to 90% to the lower reaches of the watershed, has a considerably low utility efficiency. The water accessibility could be a quantitative measure of water scarcity in the mountains. It can be used effectively for emergency water shortage planning and water resource management. In the present study, Dongchuan District, a typical county in the Hengduan Mountains in Yunnan province of Southwest China, was selected as the study area, and the minimal cumulative resistance(MCR) model was used to simulate the least-cost path(LCP) from 1255 point features of natural villages, as well as 12,368 dryfield centroids, to their respective surrounding river systems, which serve as a source for emergency drinking water and irrigation during droughts. The average length of the LCP for each administrative village was calculated to represent the accessibility to water sources for agricultural production and daily life in these mountain villages. The distribution of population and dryfields, as well as other geographic elements, were analyzed to classify the degree of water scarcity in these villages. The results indicate that the area facing the highest risk of water shortage for agricultural irrigation is located in northern Dongchuan, in particular along the two sides of the Xiaojiang Valley, and that the area with the highest risk of water shortage for daily life needs is located along the Xiaojiang Valley.     

基于Cubist模型树的城市不透水面百分比遥感估算模型

不透水面是城市区域中一种典型的土地覆盖类型,是衡量城市环境质量和城市化水平的重要标志之一。与传统基于像元级的遥感研究方法相比,不透水面百分比（Impervious Surface Percent,ISP）的估算可以进入像元内部,获得更准确的城市信息。本文应用Cubist模型树,对Landsat TM的原始波段变量（除热红外波段）,建立ISP估算的基础模型（Base Cubist-ISP）。通过基于模型树的集成学习优化算法和加入相邻时相影像的波段变量中值,以削弱噪声的影响。然后,优选热红外波段和各种衍生变量,并进行属性精简,继而应用集成学习算法得到的参数和精简后的变量建立ISP估算的优化模型（Optimal Cubist-ISP）。对广东省广州市海珠区的实验结果表明,Optimal Cubist-ISP模型估算不透水面的整体均方根误差（RMSE）为12.98%,决定系数（R²）为0.90,精度明显优于Base Cubist-ISP模型,RMSE降低约5.03%,ISP在透水面区域被高估和高密度不透水面区域被低估的现象得到改善。本文提出的基于Cubist模型树建立ISP遥感估算的模型及优化方法可以适用于城市区ISP的提取。     

顾及人群集聚和情绪强度的城市综合活力评价及影响因素

城市活力的科学定量评估,能够为城市规划和协调发展提供重要依据,针对城市活力容易忽略居民情感的现状,本文选用百度热力图数据与微博情感分析结果,分别衡量人群集聚强度和情绪强度,并结合TOPSIS方法,提出一种顾及人群集聚和情绪强度的综合活力评估框架。从城市物理环境、经济环境和生态环境3个维度,选择8个关键影响因子,结合地理探测器空间分析方法,探讨影响因子对城市活力空间异质性的影响。结果表明：① 融合人群集聚强度和情绪强度的综合活力评估方法,能够较好反映城市活力空间分异格局;通过对典型样本区域分析,验证了本文提出的城市综合活力评价框架的有效性;② 城市POI密度对城市综合活力的解释力最显著,而植被覆盖度因子对城市综合活力的解释力最弱;但是植被覆盖度因子与其他因子的交互作用,对于城市活力空间异质性的影响力提升最为显著,表明植被覆盖度因子并不是直接作用于城市活力的空间异质性,而是通过耦合空间可达性、POI密度以及建筑密度等影响因子,间接影响城市综合活力的空间分异。     

基于网络地图的多交通模式医疗设施可达性分析

由于城镇化发展迅速、居民对医疗服务需求的提高,使得城市人口密集区域医疗服务压力增大,因此进行医疗服务设施可达性分析具有重要意义。本文以福州市主城区为例,医疗数据来自于福州市卫健委,利用爬虫技术获得小区户数进行人口估算,基于早、中、晚各时段的实时路况信息,计算居民点到医疗服务设施的最优路径旅行时间,并绘制医疗服务等时区,利用高斯距离衰减函数改进的两步移动搜法（Gaussian-Two Step Floating Catchment Area Method, Ga-2SFCA）,考虑出行模式,对福州市主城区二级以上医院进行可达性分析。结果表明：(1)驾车出行模式下,医疗服务覆盖率和居民就医可达性要明显好于公共交通出行模式;(2)驾车出行模式下,医疗服务可达性受出行时段影响较大;而公共交通出行方式下,各时段差异较小;(3)随着时间阈值的增大,居民点的医疗服务可达性等级提升,且高等级可达性范围逐渐扩大;(4)驾车模式下的医疗可达性空间分布和道路保持一致,呈现出“环线层次”的现象;而公共交通模式下的可达性空间分布由于受到城市公交微循环系统的影响,呈现出“轴向扩张”的现象。     

Impact of land expropriation on farmers' livelihoods in the mountainous and hilly regions of Sichuan,China

Research on the sustainable livelihoods of rural households is of great significance to mitigating rural poverty and reasonable land expropriation policy helps to realize better livelihoods and sustainable development. Scholars have conducted considerable research on the relationships between land expropriation and farmers. livelihoods. However,few quantitative studies have used the characteristics of villages as control variables to systematically analyze the impact of land expropriation on farmers.livelihood capital and strategy in the mountainous and hilly regions of China. This study uses the Partial Least Squares-Structural Equation Model(PLS-SEM)to systematically explain the impact of landexpropriation on farmers. livelihoods in mountainous and hilly regions of Sichuan in 2013, with the characteristics of the village including income,accessibility and terrain as control variables. The analysis uses both representative sample data of 240 rural households and spatial data calculated using a 30 m Digital Elevation Model(DEM) by Geographic Information System(GIS). Results are as follows:(1)The land expropriation characteristics are negatively affected by village accessibility. Villages with worse accessibility tend to have fewer land parcels and less land area expropriated. Additionally, land that is expropriated from inaccessible villages tends to receive less compensation.(2) Natural capital is negatively affected by number and area of land expropriation. Natural capital is not only directly affected by village accessibility, but also indirectlyaffected by village accessibility through the mediating effect of the number and area of land parcels expropriated.(3) Physical capital is positively affected by compensation for land expropriation, and negatively affected by village accessibility through compensation for land expropriation. The worse a village.s accessibility/location is, the less compensation it will receive for land expropriation,resulting in lower physical capital.(4) Financial capital is negatively affected by village accessibility indirectly through compensation for land expropriation. The better the village.s accessibility is,the greater is its compensation for land expropriation and, hence, the greater is its financial capital.(5)Social capital is directly and negatively affected by the number and area of land parcels expropriated, and is indirectly and positively affected by village accessibility through the number and area of land parcels expropriated. This study enhances our understanding of the characteristics of land expropriation and rural households. livelihood as well as the impact of land expropriation on rural households. livelihood. These findings provide reference for the formulation of proper policies related to land expropriation and the improvement of rural households. livelihoods in the mountainous and hilly regions of China.     

Spatial Accessibility to Shopping Malls in Nanjing,China: Comparative Analysis with Multiple Transportation Modes

This study attempts to evaluate the rationality of the spatial layout of shopping malls and identify the urban area with poor shopping mall accessibility, as well as to provide a reference and assist decision-making for planning. Using the case of Nanjing, China, this study developed a method to evaluate the accessibility of shopping malls by three transport modes(car traffic, public transit, and bicycle). Specially, we divide the urban space into a regular hexagonal grid and harvest the total travel time from each of 7204 hexagon centroids to each of 129 shopping malls using the Baidu Internet map. The door-to-door travel time approach is used to evaluate all travel stages(walking, waiting, transfer, and transportation) based on travel time calculations. We further divide the shopping malls into two levels(super-regional and regional) based on the Dianping App's information and develop the indicator of accessibility to shopping malls: the number of shopping malls within tolerance time thresholds and apply the closest facility and cumulative opportunities methods to measure accessibility scores. The results show that the accessibility estimations vary greatly with transport modes. The accessibility of shopping malls presents a concentric ring trend centered on the city center under the car traffic and bicycle modes. And public transit accessibility tends to axially extend due to the topology of bus routes and metro lines. In particular, we observe that the accessibility of shopping malls in Nanjing has an uneven spatial distribution pattern, with high accessibility values in the central urban areas and lots of underserved areas in urban fringe regions. Based on the accessibility measurements, we finally map the poor accessibility area and propose corresponding implications for urban planning.     

广东区域公路网络的城市通达性度量及空间特征分析

以广东省各地级市辖区和县域中心点为公路网络节点,通过对各节点城市的中心职能强度指数分析及其相互之间在公路网络的连通关系为基础,综合应用日通达性模型、通达性度量模型和潜力模型,对广东区域通达性进行分析,并以不同指标和度量模型的广东区域通达性的空间格局和分异特征进行GIS分析。结果表明,广东省的城市通达性空间格局,总体上呈现出一种圈层空间分布结构,以广州市核心区域和佛山市中心区域(禅城区和顺德区)为中心向外辐射。其影响通达性区域差异的主要因素包括城市经济水平、交通条件和地理区位等,其中,在交通条件影响下的通达性区域差异,在空间上表现出高等级公路网络指向性。研究结果反映出广东省省域公路网络通达性的空间分异规律,诠释了城市的经济发展水平、交通条件和地理区位对通达性的综合作用,从而为广东省城镇网络体系发育、区域经济协调及珠三角产业转移规划优化提供科学支撑。     

武汉市轨道交通站点周边地区职住平衡与建成环境的关系研究

轨道交通的快速发展提高了通勤可达性,也被认为在重构城市职住空间格局进而影响职住关系方面产生重大影响.探究城市轨道交通站点周边地区(下称轨交站点周边地区)就地职住平衡与建成环境的关系,对于通过建成环境设计促进站点尺度职住"微平衡",调节由于轨道交通所导致的区域性职住分离意义重大.本文基于武汉市189个轨交站点刷卡数据,首...