彩色交通栅格地图中城市道路与公路的自动获取

发布时间：2018-01-23 01:44

本文关键词： 彩色交通栅格地图数字地图道路自动获取颜色自动获取颜色聚类噪声消除特征匹配　出处：《中国科学技术大学》2013年博士论文　论文类型：学位论文

【摘要】：随着交通地理信息系统的发展,快速建立交通数字地图日趋重要。道路矢量数据是交通矢量地图的基础数据,彩色交通栅格地图是道路矢量数据的一个重要信息源。从彩色栅格地图中自动获取道路是快速、高效、低成本地自动生成道路矢量数据的重要途径。人工识别彩色交通栅格地图中道路较容易,利用计算机自动获取道路则是一个挑战性的课题。本论文以彩色城市交通栅格地图和彩色公路交通栅格地图为研究对象,综合利用两类彩色交通栅格地图的特征,重点对地图要素的颜色自动获取、噪声消除方法、点状符号自动获取几个方面进行研究。利用本论文提出的方法,实现了一个彩色交通栅格地图道路自动获取系统。本论文的主要研究成果与创新点： (1)综合利用彩色交通栅格地图中背景、道路与噪声的特征,提出背景颜色自动获取的同色块检测法、城市道路颜色自动获取的游程长度分析法、公路颜色自动获取的噪声消除法和公路干道颜色自动获取的邻域法,应用这些方法准确地获取了彩色交通栅格地图中背景与道路的颜色,解决背景与道路的颜色获取需人工干预的问题。 (2)提出并综合线状噪声、块状噪声、梯度特征法等多种噪声消除方法,解决彩色栅格交通地图中道路颜色已知和未知两种情况下的噪声消除问题,获得了比现有噪声消除方法更好的噪声消除效果,克服了现有的噪声消除方法在背景与道路的颜色聚类中消除大量噪声的不足。 (3)利用彩色公路交通栅格地图中县城符号的形状和颜色分布特征,提出与实现一种县城符号自动获取的特征匹配法,应用该方法从不同颜色的彩色公路交通栅格地图中准确地提取出县城符号,克服模板匹配法对彩色公路栅格地图的颜色变化敏感、阈值难以选取和计算量大的不足,为实现公路干道颜色自动获取的邻域法奠定基础。本论文先结合县城符号的图例及其Laplace边缘图像选取两组特征点,再依据两组特征点的统计特征识别县城符号,最后通过实验验证特征匹配法的有效性和优越性。 (4)综合上述方法,实现彩色交通栅格地图中城市道路与公路的自动获取,设计与实现彩色交通栅格地图城市道路与公路自动获取系统。先实现数字图像处理系统的基本功能,再实现彩色交通栅格地图中城市道路与公路的自动获取,最后通过实验验证城市道路与公路自动获取方法的有效性与优越性。
[Abstract]:With the development of traffic geographic information system, it is more and more important to set up traffic digital map quickly. Road vector data is the basic data of traffic vector map. Color traffic grid map is an important information source of road vector data. It is fast and efficient to automatically obtain the road from color grid map. It is an important way to generate road vector data automatically at low cost. It is easy to manually identify the road in the color traffic grid map, and it is a challenging task to obtain the road automatically by computer. This paper takes the color urban traffic grid map and the color highway traffic grid map as the research object, synthetically uses the two kinds of color traffic grid map's characteristic, emphatically obtains the map essential factor color automatically. In this paper, a color traffic grid map road automatic acquisition system is implemented by using the method proposed in this paper. The main research results and innovations of this thesis are as follows: 1) based on the features of background, road and noise in color traffic grid map, this paper puts forward a method of automatic background color acquisition by the same color block detection method, and a run length analysis method of city road color automatic acquisition method. The noise elimination method of automatic acquisition of road color and the neighborhood method of automatic acquisition of road color are used to accurately obtain the background and the color of the road in the color traffic grid map. To solve the background and road color acquisition needs manual intervention. In order to solve the problem of road noise elimination in color grid traffic map, several noise elimination methods, such as linear noise, block noise and gradient feature method, are proposed and synthesized. A better noise elimination effect is obtained than the existing noise elimination methods, and the shortcomings of the existing noise elimination methods in background and road color clustering are overcome. Based on the shape and color distribution features of county town symbols in the color highway traffic grid map, a feature matching method for automatic acquisition of county town symbols is proposed and implemented. The method is used to extract the county town symbols accurately from the color road traffic grid map of different colors, and to overcome the sensitivity of the template matching method to the color change of the color highway grid map. It is difficult to select the threshold value and the shortage of large amount of calculation. In order to achieve the road color automatic acquisition of the neighborhood method to lay the foundation. Firstly, this paper combines the legend of county town symbol and Laplace edge image to select two groups of feature points. Based on the statistical features of the two groups of feature points, the county seat symbols are identified. Finally, the effectiveness and superiority of the feature matching method are verified by experiments. The above method is integrated to realize the automatic acquisition of urban roads and highways in the color traffic grid map. Design and implementation of color traffic grid map urban road and highway automatic acquisition system. First, realize the basic function of digital image processing system, and then realize the automatic acquisition of urban road and highway in color traffic grid map. Finally, the effectiveness and superiority of the automatic acquisition method of urban roads and highways are verified by experiments.
【学位授予单位】：中国科学技术大学
【学位级别】：博士
【学位授予年份】：2013
【分类号】：U495;P208

【参考文献】