彩色扫描地图沙地符号去除算法研究

发布时间：2018-03-03 13:25

本文选题：沙地符号　切入点：MapGIS平台　出处：《西安电子科技大学》2013年硕士论文　论文类型：学位论文

【摘要】：数字地图矢量化是信息领域一个热门的研究课题，也是建立地理信息系统的一个必要步骤。沙漠地图矢量化是地图矢量化中的重要部分，该研究主要关注于沙漠地区线性元素的信息。但是沙漠地图中的沙地类型符号严重影响了线性信息的矢量化效率和结果。针对上述问题，本文提出了两种沙漠符号去除新算法，并将其集成到MapGIS平台下。本文主要的研究工作和创新点如下： 1.针对沙漠地图图像中出现的沙地类型符号的特点、相互位置关系，通过将原始地图图像转换为二值图像，再经过细化处理后，使用连通域阈值过滤以及像素跟踪算法完成对平沙地符号和窝状沙地符号的去除。同时，，算法最大程度保留了图像中线性元素本来的特征，如方向和形状，维持了信息的完整性。 2.在研究算法理论的基础上，将理论与实际应用相结合，将该算法集成到MapGIS遥感测绘平台中，并且通过工具栏按钮的形式向用户提供。通过简单的参数设置，用户就能够使用该方法有针对性进行图像处理工作。通过算法理论的不断完善和在实际工作的检验，该算法获得良好效果，很大程度上提高了矢量化工作效率。
[Abstract]:Digital map vectorization is a hot research topic in the field of information, and it is also a necessary step in the establishment of geographic information system. Desert map vectorization is an important part of map vectorization. This study focuses on the information of linear elements in desert area, but the symbol of sand type in desert map seriously affects the vectorization efficiency and result of linear information. To solve the above problems, two new desert symbol removal algorithms are proposed and integrated into the MapGIS platform. The main research work and innovation of this paper are as follows:. 1. According to the characteristics of sand land type symbols in desert map images, the original map images are transformed into binary images, and then refined after processing. The connected domain threshold filtering and pixel tracking algorithm are used to remove the symbols of flat sand and nest sand. At the same time, the original features of linear elements in the image, such as direction and shape, are preserved to the maximum extent. The integrity of the information was maintained. 2. Based on the research of the algorithm theory, the algorithm is integrated into the MapGIS remote sensing surveying and mapping platform by combining the theory with the practical application, and it is provided to the user through the form of the toolbar button. The user can use the method to deal with the image. Through the continuous improvement of the algorithm theory and the test of practical work, the algorithm obtains good results, and greatly improves the efficiency of vectorization.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：P208;TP391.41

【参考文献】