基于遥感图像的大规模城市建模中的建筑物轮廓提取-光学工程专业论文.docxVIP

万方数据 万方数据 摘 要 I 摘 要 3D 数字城市是视景仿真以及数字地球的基础。实现城市数字化的关键是进行 快速有效的城市建模。但多数建模方法都需要很多先验信息甚至大量人力，并且 手工建模很难处理大规模分布的城市元素，所以仅依靠手工建模不容易满足大规 模城市的高效建模。 本文采用了一种基于遥感图像的、无成本的、灵活的自动 3D 数字城市建模方 法，其主要内容包括图像分割、图像形态学运算、建筑物轮廓提取、3D 模型重建、 纹理贴图等，可有效解决手工建模的不足。 在上述基于图像的三维重建方法中，建筑物轮廓的识别与提取决定着最后的 重建效果，而图像的分割与形态学运算直接影响着建筑物轮廓的提取。文章重点 论述了基于遥感图像的建筑物轮廓提取过程及方法，并研究了图像分割和图像形 态学对最后结果的影响。对于遥感图像中建筑物轮廓的提取，本文采用了区域特 征与边缘信息相结合的方法：首先对图像采用 Mean Shift 多尺度分割算法进行预 分割，随后再用 Otsu 阈值算法对其进行后续分割，初步清除图像中的大部非建筑 物部分；最后使用形态学运算对分割结果进行优化，有效地将与建筑屋顶颜色特 征及光谱特征都相似的道路分离出去。文章最后对基于建筑物轮廓的三维重建过 程进行了简要介绍。 实验结果表明：本文所用方法可达到令人满意的建筑物提取效果，实用价值 较高；易于理解与实现也是本文所用方法的优点。 关键字： 遥感图像分割 图像形态学运算 建筑物轮廓提取 城市三维重建 ABSTRACT III ABSTRACT 3D digital city is the basis of the visual simulation and digital earth. Fast and efficient city modeling technology is the key to implement digital city. But most of the modeling method requires a lot of prior information and even a large number of manpower. Dealing with large-scale distribution of urban elements is of great difficulty. Only relying on manual modeling is not easy to satisfy the efficient modeling of Large-scale urban. A low-cost、flexible and automatic modeling method of 3D digital city , which is based on remote sensing image and can solve the shortcomings of manual modeling effectively is adopted in this paper. Its main contents include image segmentation、 image morphology operation、building contour extraction、reconstruction of 3D model、 texture mapping, which can effectively solve the shortage of manual modeling. In the 3D reconstruction method based on image above, the outline of buildings determines the final reconstruction effect, while the image segmentation and morphological operation directly affects the building outline extraction. The article mainly discusses the process and method of building contour extraction based on remote sensing image. At the same time, image segmentation and image morphology are studied for the extraction results. For building contour extraction in remote sensing image,