• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文学位 >Understand the social impact of green---evaluation of the impacts of urban vegetation on neighborhood crime.
【24h】

Understand the social impact of green---evaluation of the impacts of urban vegetation on neighborhood crime.

机译：了解绿色的社会影响-评估城市植被对社区犯罪的影响。

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

This dissertation addresses one of the fundamental issues in landscape architecture, environmental planning as well as the general field of environment and behavior: the relationship between vegetation settings and neighborhood safety. Being built upon a critical debate on whether vegetation is actually helpful or harmful to crime prevention, this piece of research seeks to answer how different types of vegetation affect various categories of crime in neighborhoods with varied socio-economic and environmental characteristics. It has a firm grounding in a group of interdisciplinary theories, including environmental psychology, crime prevention through environmental design (CPTED), and neighborhood criminology. Through synthesizing these theories, an innovative theoretical framework that integrates vegetation factors with traditional crime factors identified by major schools of thought of neighborhood criminology is constructed to comprehensively explain the incidence of neighborhood crime. More specifically, the study seeks to answer how the existence of various types of vegetation settings is related the occurrence of neighborhood crime after removing the effects of major crime factors as identified by neighborhood criminology literature. To achieve these objectives, a data rich environment is built for the investigation of the issue through collecting a rich set of geospatial and attribute data on all relevant factors, including vegetation settings and traditional crime factors, for the study area, i.e. Oakland, California.;With the huge raw dataset at hand, the primary task of research is to transform them into a uniform format through remote sensing image processing and GIS modeling. The data transformation process is therefore in two phases. The first one was to extract detailed, individual-tree level vegetation cover and vegetation height from very high resolution remote sensing images. A comprehensive approach, which consists of image registration, segmentation, object-based image classification, ground truth sampling, accuracy evaluation, and geospatial modeling, is developed for this purpose. A shadow modeling method comprised of geometric modeling, spatial decomposition, geo-processing and other GIS methods is then applied to extract vegetation height from detail image objects of vegetation and shadow. Next, the study further models the view-blocking effects of vegetation settings through a rule-based hierarchical evaluation system. Lastly the accuracy of the approaches is checked with in situ data collected with GPS instruments. Accuracy assessment indicates that the approaches are successful in measuring detailed vegetation distribution and their view-blocking effects.;The second phase of data processing focuses on spatial data integration and modeling. In this phase we transform data from multiple sources and in multiple formats to data in uniform format and consistent quality with state of the art geospatial methods and automated GIS procedures. GIS methods applied include geo-coding, spatial sampling, areal interpolation, network routing, geospatial statistics, etc. As outcome, a new geo-database containing uniform data for over two hundred variables is generated for each of the sampled neighborhoods in Oakland.;To analyze the entire dataset, the research applies multiple regressions and multivariate analyses to answer the proposed research questions, i.e. whether there is a significant relationship between vegetation settings and neighborhood crime, after having controlled the effects of other crime factors. Results of this research prove that the integration of geospatial modeling and advanced statistical analyses is of central importance to ensure the explanation of up to 97% of Part I neighborhood crime. In light of their effects on crime, vegetation settings are found to have a positive relationship with property crime, while a negative relationship with most violent crime. More high view-blocking vegetation settings, especially in public space, are significantly correlated with more violent crime but less property crime. In summary, the relationships between vegetation settings and crime are more complex than that were reported in the literature. How vegetation settings affect neighborhood safety is not only a matter of demographic and socioeconomic status of people, planning and management of urban space, and the construction of landscape settings, but also determined by the mechanisms, through which different types of crime occur in varied social and physical context. Therefore, for various categories of crime, the impacts of landscape settings could be totally different. (Abstract shortened by UMI.)
机译：本文主要研究景观设计，环境规划以及环境与行为的一般领域中的基本问题之一：植被环境与社区安全之间的关系。这项研究以关于植被实际上对预防犯罪有帮助还是有害的激烈辩论为基础，旨在探讨具有不同社会经济和环境特征的社区中不同类型的植被如何影响各种犯罪。它在一组跨学科理论中具有牢固的基础，包括环境心理学，通过环境设计预防犯罪（CPTED）和社区犯罪学。通过综合这些理论，构建了一个创新的理论框架，将植被因素与社区犯罪学主要学派确定的传统犯罪因素相结合，以全面解释社区犯罪的发生。更具体地说，该研究试图回答在消除主要犯罪因素的影响之后，各种类型的植被环境如何与邻里犯罪的发生相关联，这些犯罪因素是由邻里犯罪学文献确定的。为了实现这些目标，通过为研究区域（即加利福尼亚州奥克兰）收集有关所有相关因素（包括植被设置和传统犯罪因素）的丰富地理空间和属性数据，构建了一个用于调查问题的数据丰富的环境。 ；有了庞大的原始数据集，研究的首要任务是通过遥感图像处理和GIS建模将它们转换为统一格式。因此，数据转换过程分为两个阶段。第一个是从非常高分辨率的遥感图像中提取详细的单棵树植被覆盖度和植被高度。为此，开发了一种综合方法，该方法包括图像配准，分割，基于对象的图像分类，地面真实采样，准确性评估和地理空间建​​模。然后应用由几何建模，空间分解，地理处理和其他GIS方法组成的阴影建模方法，从植被和阴影的细节图像对象中提取植被高度。接下来，该研究通过基于规则的层次评估系统进一步模拟植被设置的视线遮挡效果。最后，使用GPS仪器收集的现场数据检查方法的准确性。准确性评估表明该方法成功地测量了详细的植被分布及其遮挡效果。;数据处理的第二阶段着重于空间数据集成和建模。在此阶段，我们将数据从多种来源以多种格式转换为统一格式的数据，并通过最先进的地理空间方法和自动化的GIS程序来保证质量的一致性。所应用的GIS方法包括地理编码，空间采样，区域插值，网络路由，地理空间统计等。结果，为奥克兰的每个采样邻域生成了一个包含200多个变量的统一数据的新地理数据库。为了分析整个数据集，该研究应用多元回归和多元分析来回答所提出的研究问题，即在控制了其他犯罪因素的影响之后，植被环境与邻里犯罪之间是否存在显着关系。这项研究的结果证明，地理空间建​​模与高级统计分析的集成对于确保对多达97％的第一部分邻里犯罪做出解释至关重要。根据其对犯罪的影响，发现植被环境与财产犯罪有正相关关系，而与大多数暴力犯罪有负相关关系。较高的遮挡植被设置，尤其是在公共场所，与暴力犯罪增加但财产犯罪减少明显相关。总之，植被状况和犯罪之间的关系比文献中报道的更为复杂。植被如何影响邻里安全不仅取决于人的人口和社会经济地位，城市空间的规划和管理以及景观设置，还取决于各种机制，在不同的社会中，不同类型的犯罪发生和物理环境。因此，对于各种犯罪，景观环境的影响可能完全不同。 （摘要由UMI缩短。）

著录项

  • 作者

    Li, Weimin.;

  • 作者单位

    University of California, Berkeley.;

  • 授予单位 University of California, Berkeley.;
  • 学科 Geography.;Sociology Criminology and Penology.;Landscape Architecture.;Environmental Management.;Urban and Regional Planning.
  • 学位 Ph.D.
  • 年度 2008
  • 页码 456 p.
  • 总页数 456
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类 自然地理学;地下建筑;法学各部门;区域规划、城乡规划;
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号