项目名称: 基于LiDAR数据的非栅格化道路矢量提取及融合高分影像的路网探测与优化
项目编号: No.41501454
项目类型: 青年科学基金项目
立项/批准年度: 2016
项目学科: 天文学、地球科学
项目作者: 李怡静
作者单位: 南昌大学
项目金额: 20万元
中文摘要: 遥感数据的目标自动识别与提取是国家基础地理信息建设与更新的重要手段,然而城市道路多样的表现形式及复杂的上下文关系加大了对其自动识别与提取的难度。LiDAR数据与高分辨率遥感影像各自特征的优势与局限,为道路提取研究带来新机遇和挑战。本项目针对LiDAR数据离散分布不均的结构特征和高分辨率遥感影像的清晰边缘及丰富语义,研究多信息融合的城市道路网提取方法,内容包括(1)基于LiDAR数据的道路骨架线提取策略,将多特征聚类与形态感知结合,非栅格化提取道路骨架线,以降低处理步骤和参数,增强方法普适性;(2)基于高分辨率遥感影像的顾及路面上下文的道路探测算法,利用主方向显著性分析实现道路探测,以提高结果的正确性;(3)融合两种遥感数据源多重信息的道路网智能优化方法,实现拓扑关系完整的路网提取。该研究将为城市地区复杂场景下的道路网自动提取提供新的解决方案。
中文关键词: 道路提取;机载激光雷达;目标识别;遥感影像
英文摘要: Road network is one of the most important information of basic geography. However, roads extraction is a very difficult task due to the occlusions and shadows of the contextual objects and the complex pattern of the roads. LiDAR data and high-resolution remote sensing images have their own advantages and limitations,which bring new research directions and challenges for road extraction. Combining LiDAR data and imagery can improve the performance of automatic road extraction. Therefore, a proper fusion methodology to achieve a better outcome is explored in this research: (1) For the irregular distribution of LiDAR data, a new method of automatic road extraction without image interpolation and prior model is studied. (2) A road detection algorithm from high-resolution remote sensing image supported by contexts objects and road edge is explored to improve the correctness of road center lines. (3) A fusion method based on multi-information from LiDAR data and imagery is researched to achieve complete road networks. The research will provide a new solution for the automatically extraction of road networks from remote data in urban areas.
英文关键词: Road extraction;LiDAR;Target recognition;Remote sensing image