图残差神经网络支持下的建筑物群组模式分类

建筑物作为城市中的重要地物,分析其群组模式对地图综合、导航定位、市政规划等具有重要作用。建筑物群组模式分析目前主要有基于规则的方法和基于机器学习的方法两种。基于规则的方法和基于传统机器学习分类器的方法均需要大量的人工处理过程。近年来兴起的深度学习特别是图卷积神经网络前期无需人工处理,因此提高了建筑物群组模式分析的自动化程度。传统的图卷积神经网络模型在训练深层网络时易出现退化问题,提取深层特征困难。为解决此问题,本文引入了图残差神经网络模型用于建筑物群组的模式分类。首先使用道路和河流等作为约束条件,利用K-means方法对建筑物进行聚类;然后根据Bertin视觉变量计算对应的建筑物特征指标,在每个建筑物群组中以建筑物质心为节点,连接节点的最小生成树作为边,构建建筑物群组图结构;最后将得到的图结构数据输入图残差神经网络进行训练,得到规则和不规则两种建筑物群组模式。试验结果表明,该模型较好地解决了传统图卷积神经网络模型的退化问题,并取得了更高的精度。     

基于植被分布特征的森林景观建模方法

在虚拟地理环境中如何快速地生成具有特定空间格局的森林景观是一个具有挑战的问题。现有的研究主要以模拟或随机的方式确定树木的位置,但难以反映每个树种的分布特征。本文提出了一种基于图的中性景观模型,根据指定的景观指数生成特定组成结构和空间配置的森林景观。模型将格网随机划分为若干不规则的区域,并将其转化为对应的区域邻接图,通过源节点的受限生长生成特定大小和形状的图斑。相比其他中性景观模型,该模型在可控性和灵活性上具有优势,能对每种类型的景观特征分别进行表达。最后根据模型生成的森林景观确定树木的位置与种类。本方法不仅能为植被场景的可视化提供合理的植被分布,而且能反映森林景观的空间分布特征,为虚拟森林环境中的林业规划决策以及植被分布规律认知提供一定的支持。     

基于Delaunay三角网模型的等高线地形特征提取及谷地树结构化组织

基于Delaunay三角网模型在矢量等高线基础上通过弯曲特征分析，给出了一种地形结构特征提取及谷地树组织的方法，与传统的基于DEM或TIN数据结构提取地形特征的方法相比，该方法不仅得到了谷地系统在平面图上的拓扑结构，还获得了河谷在空间分布上的汇水范围，依托矢量线、多边形的几何关系判断，增强了谷地树组织中的结构化，避免了DEM方法中的噪音干扰，方法的核心是Delaunay三角网在单根等高线弯曲深度表达上的层次结构识别。基于系统地实验详细讨论了3种树的结构化组织：等高线弯曲特征嵌套结构表达的二叉树、谷地系统拓扑关系表达的平面结构树、水文意义上主支流河谷表达的语义层次树。     

基于Delaunay三角网模型的等高线地形特征提取及谷地树结构化组织

基于Delaunay三角网模型在矢量等高线基础上通过弯曲特征分析，给出了一种地形结构特征提取及谷地树组织的方法，与传统的基于DEM或TIN数据结构提取地形特征的方法相比，该方法不仅得到了谷地系统在平面图上的拓扑结构，还获得了河谷在空间分布上的汇水范围，依托矢量线、多边形的几何关系判断，增强了谷地树组织中的结构化，避免了DEM方法中的噪音干扰，方法的核心是Delaunay三角网在单根等高线弯曲深度表达上的层次结构识别。基于系统地实验详细讨论了3种树的结构化组织：等高线弯曲特征嵌套结构表达的二叉树、谷地系统拓     

On the spatial distribution of buildings for map generalization

Information on spatial distribution of buildings must be explored as part of the process of map generalization. A new approach is proposed in this article, which combines building classification and clustering to enable the detection of class differences within a pattern, as well as patterns within a class. To do this, an analysis of existing parameters describing building characteristics is performed via principal component analysis (PCA), and four major parameters (i.e. convex hull area, IPQ compactness, number of edges, and smallest minimum bounding rectangle orientation) are selected for further classification based on similarities between building characteristics. A building clustering method based on minimum spanning tree (MST) considering rivers and roads is then applied. Theory and experiments show that use of a relative neighbor graph (RNG) is more effective in detecting linear building patterns than either a nearest neighbor graph (NNG), an MST, or a Gabriel graph (GssG). Building classification and clustering are therefore conducted separately using experimental data extracted from OpenStreetMap (OSM), and linear patterns are then recognized within resultant clusters. Experimental results show that the approach proposed in this article is both reasonable and efficient for mining information on the spatial distribution of buildings for map generalization.     

数学形态学和模式识别在建筑物多边形化简中的应用

针对居民地图形化简的一个方面--建筑物多边形的化简,提出一种在与地图比例尺相关的动态栅格和矢量数据相结合的数据模型支持下,综合利用数学形态学和神经网络支持下的模式识别的化简方法.在Visual C 环境下实现基于此方法的系统并进行实验,实验结果说明此方法在保持街区的形态特征上效果明显.这种方法将制图综合知识融入图形化简操作之中,是自动制图综合智能化的一次新的尝试.     

基于结构特征的地图上带晕线多边形的识别

在建立GIS的过程中,从现有的各种比例尺的地形图上获取地理信息是必不可少的工作。带晕线多边形是基本比例尺地形图上最主要的地物要素类别之一,文中在地图二值图像的基础上,根据带晕线多边形的结构特征,利用一系列图像变换技术,首次成功地实现了带晕线多边形的自动识别与提取。提出并实现了用于带晕线类多边形影像分割的4方向RLS变换方法,详细讨论了收缩变换、扩张变换、细化、开图形删除和带晕线类多边形的识别等算法。实验表明,利用文中所述方法进行带晕线多边形识别与提取的正确率可达96%左右。     

Grid pattern recognition in road networks using the C4.5 algorithm

Pattern recognition in road networks can be used for different applications, including spatiotemporal data mining, automated map generalization, data matching of different levels of detail, and other important research topics. Grid patterns are a common pattern type. This paper proposes and implements a method for grid pattern recognition based on the idea of mesh classification through a supervised learning process. To train the classifier, training datasets are selected from worldwide city samples with different cultural, historical, and geographical environments. Meshes are subsequently labeled as composing or noncomposing grids by participants in an experiment, and the mesh measures are defined while accounting for the mesh’s individual characteristics and spatial context. The classifier is generated using the C4.5 algorithm. The accuracy of the classifier is evaluated using Kappa statistics and the overall rate of correctness. The average Kappa value is approximately 0.74, which corresponds to a total accuracy of 87.5%. Additionally, the rationality of the classifier is evaluated in an interpretation step. Two other existing grid pattern recognition methods were also tested on the datasets, and comparison results indicate that our approach is effective in identifying grid patterns in road networks.     

Estimation of sub-catchment area parameters for Storm Water Management Model (SWMM) using geo-informatics

This study presents a remote sensing and geographic information systems-based approach for using US EPA’s Storm Water Management Model (SWMM) in urban environment. Cartosat-1 PAN + IRS-P6 LISS-IV merged product was used to map land cover in part of Surat city at 1:10,000 scale. Cartosat-1 stereo pair was used for deriving digital elevation model of the study area. Geo-informatics-based methods were developed for delineation of sub-catchment areas, assignment of sub-catchment outlets and estimation of characteristic width. It was observed that 59% of the developed area in the study region was directly or indirectly connected to the storm water drainage network. Furthermore, dynamic rainfall-runoff simulation on three-day rainfall indicated that the average runoff coefficient on the urbanized sub-catchment areas which were directly connected to the drainage network was 0.92 as against 0.88 on those urbanized sub-catchments without having direct access to storm water drainage.     

城市建筑群网格模式的图论识别方法

建筑群空间分布模式体现了城市的物质形式及其与社会经济功能之间的关系,反应了城市的空间结构特征,对于制图综合和多尺度表达等具有重要意义。结合国内外对该问题的研究,提出了基于图论的建筑群网格模式识别方法。首先分析研究了网格模式的认知特征和定义。然后利用Delaunay三角网构建建筑群的邻近关系,生成邻近图;从Gestalt视觉准则出发,基于三角剖分模型建立视觉距离;考虑直线模式的直线性、紧凑性等约束条件识别出交叉的多连通直线模式。最后对直线模式建立相交图和方向关系图,通过求解极大完全子图、连接、相交和后期修建等图运算,实现网格模式的识别。实验表明该方法能够识别出明显网格模式,其识别结果符合人类空间认知特点。     

News Items

The process of reading a topographic map requires users to recognize and learn the cartographic symbols of the key (or legend) while interpreting the territory as depicted on the map at a given level of abstraction (the form and nature of features, their saliency and relationships). We present the results of an empirical user study that aims to identify and assess the main graphical characteristics that are used by map users to recognize the design principles which constitute the topographic mapping style adopted by IGN (Institut National de l'Information Géographique et Forestière), France. Our results suggest that 91% of the participants were able to recognize an IGN-France topographic map amongst other topographic map products. We also determine which graphical characteristics play a role in the recognition of this cartographic style, either by visual memory or by visual perception, and identify the representation of relief, including contour lines and shaded relief, as one of the major graphical characteristics of the topographic mapping style of IGN-France. Moreover, the participants of our study considered the representation of touristic POI (points of interest), toponymy, typography, the main roads network, the individual buildings and the forests, to be essential for stylistic recognition.     

基于BP神经网络的点状地图符号识别

点状地图符号的识别是实现扫描方式地图自动数字化的重要研究内容之一，本文首先分析介绍ＢＰ网络模型及其学习算法，然后根据ＢＰ模型本身所存在的一些不足之处，提出相应的改进措施，还以实际扫描地图图像进行相应的实验验证，最后，对网络的输入模式加以讨论，提出了三种减小网络规模的具体方法。     

一种聚集性面群中毗邻区自动识别与处理方法

毗邻化操作是具有毗邻特性的聚集性面状要素群(简称毗邻区)开展地图综合的核心内容,传统方法只给出了笼统的处理思路,难以支撑机器智能化、自动化处理。为此,本文提出一种在各种聚集性面状要素群中自动识别毗邻区及其毗邻化处理方法。首先基于Gestalt原则,提炼代表毗邻区典型特征的桥接面宽度指数(WI)、分布格局指数(DPI)、有效连接指数(ECI)和重叠度指数(OI),并进行毗邻区自动辨识;其次优化外围边界轮廓计算、毗邻化线提取等毗邻化操作关键算法,实现自动化处理;最后经江苏省某典型区域的地形图实际数据测试,检验了该自动化方法的可靠性和高效性。     

利用路段连接图识别道路网中的格网模式

提出一种基于路段连接图的格网模式识别方法.该方法以路段连接对作为研究的基本单元,以节点路段为点,路段的连接为边用路段连接图表达道路网.将在道路网中识别格网转化为在路段连接图中搜索格网回路.提出了描述路段连接对几何与连接关系的5个参量,用于筛选图中符合格网特点的节点和边.设计了图搜索的约束条件,使用广度优先遍历搜索连接关...     

江苏沿海滩涂利用模式遥感影像解译和时空变化分析

为了弄清江苏沿海滩涂利用模式分布情况,定量分析其时空变化趋势,本文基于Google Earth Engine云计算平台和Landsat卫星遥感影像数据,结合随机森林分类方法,通过机器学习建立了滩涂利用模式自动识别分类方法,样本训练后识别准确率达93％以上.利用该方法提取了2002年、2014年和2019年江苏沿海的土地...     

网络空间向量剖分法识别城市路网网格模式

将道路网络空间视为嵌在2D空间中的独立子空间,利用形态单一的线性单元剖分图结构的边,实现网络空间的栅格化;提取网格模式的典型特征,包括几何和拓扑特征,以栅格单元邻域为目标计算特征值,构建特征向量描述栅格单元,实现对象空间到特征空间的映射,构建空间向量场;基于支持向量机（support vector machine,SVM）实现网格模式分类;结合格式塔原则完善实验结果。将此方法应用于深圳市路网数据,实验结果表明能有效地识别网格模式。     

A Genetic Algorithm for Tributary Selection with Consideration of Multiple Factors

Drainage systems are important components in cartography and Geographic Information Systems (GIS), and achieve different drainage patterns based on the form and texture of their network of stream channels and tributaries due to local topography and subsurface geology. The drainage pattern can reflect the geographical characteristics of a river network to a certain extent. To preserve the drainage pattern during the generalization process, this article proposes a solution to deal with many factors, such as the tributary length and the order in river tributary selection. This leads to a multi‐objective optimization problem solved with a Genetic Algorithm. In the multi‐objective model, different weights are used to aggregate all objective functions into a fitness function. The method is applied on a case study to evaluate the importance of each factor for different types of drainage and results are compared with a manually generalized network. The result can be controlled by assigning different weights to the factors. From this work, different weight settings according to drainage patterns are proposed for the river network generalization.     

利用排水区间计算城市暴雨积水

提出了给定淹没水量条件下基于流域排水区间分布的城市暴雨积水过程计算模型。该模型针对城市暴雨积水过程的形成机理和特点提出了两个基本假设,以各积水淹没源的空间分布特征规律为依据将城市流域划分为若干个排水区间,并将排水区间作为积水淹没计算的基本单元。为了保证计算的效率,根据流域地形起伏分布状况,该模型将各个排水区间内的空间区域范围简化为一个规则几何体,并结合体积法和多种漫流计算思想模拟暴雨积水在各排水区间内部和淹没源点之间的漫延过程。实验结果表明,该模型高效、可行,对流域资料要求较低,在信息不完备条件下的城市暴雨积水模拟中具有一定的应用价值。     

Building Displacement Based on the Topological Structure

Map data at smaller scales than their source can result in spatial conflict, whereby map symbols become too close, or overlaid. Server map generalisation operators may be applied to solve this problem, including displacement. In this paper, we show how an optimisation algorithm, the snake algorithm, was used to displace multiple objects in order to resolve spatial conflicts and maintain important spatial relationships between objects during displacement. Two principles based on the snake algorithm are proposed in this paper. First, the truss structure mirroring spatial proximity relationships between buildings and between building and road is formed based on the weighted proximity graph derived from constrained Delaunay triangulations (CDT) in each map partition. In the weighted proximity graph, each connecting line is determined as a snake and as an element unit to assemble the global stiffness matrix in snake algorithm. Second, a buffer method that calculates force between a building and a road (or other linear features) or between pair of buildings is adopted in the snake algorithm. This avoids the imbalance phenomenon caused by different force calculation methods during the displacement. The feasibility of the approach is demonstrated in obtaining real geographic data. Finally, the results are cartographically usable and in particular, the spatial relationships between objects are preserved.     

直线排列建筑物群渐进式典型化方法

地图综合中,建筑物群的排列结构是需要重点考虑的因素。当不同排列的子建筑物群之间存在空间图形冲突时,这些建筑物群的综合就显得更为复杂。直线排列建筑物群的综合在大比例尺地形图上以典型化操作为主。本文提出一种相互之间存在潜在空间图形冲突的多个直线排列建筑物子群的渐进式典型化方法,渐进式地处理多个直线排列建筑物子群之间的空间图形冲突,保留建筑物群重要的直线排列结构;以建筑物表达的视觉图形约束为限制条件,自动确定典型化后的建筑物位置、形状、大小和方位。本文还研究了基于建筑物群空间邻近图的直线排列建筑物子群的自动识别方法,分析了这些直线排列之间的邻近关系和相交关系。最后,以1：5000地图上的建筑物群综合为1：25 000为试验对象,验证了所提出算法的可用性和有效性。