基于GPS数据的露天矿道路网自动提取

发布时间：2018-05-30 20:16

本文选题：露天矿 + 道路网　；参考：《煤炭学报》2017年11期

【摘要】：传统栅格法提取路网精度较低,尤其在提取露天矿道路网时,路段缺失与偏移现象更加明显。针对这一问题,摒弃了传统栅格法通过增大栅格来保证道路网连通性的处理方式,以GPS数据存在偏移这一事实为根据,假设GPS数据偏移服从正态分布,提出了通过求取GPS轨迹点在道路上的概率进行栅格初始化的方法,建立了相应模型;采用二维中值滤波方法对初始栅格数据进行修正;对查表细化算法进行改进,采用改进的细化算法对道路网栅格图像进行细化处理,得到道路网骨架信息;采用追踪法实现道路矢量化。实验表明,该方法较传统栅格方法的覆盖率提高6.43%~11.54%,错误率降低42.13%~83.02%,为道路网提取工作提供了一种有效办法,揭示了栅格大小对于路网提取结果的重要影响。
[Abstract]:The precision of traditional grid method to extract road network is low, especially in the extraction of road network in open-pit mine, the phenomenon of missing and offsetting of road sections is more obvious. In order to solve this problem, the traditional grid method is abandoned to ensure the road network connectivity by increasing the grid. Based on the fact that the GPS data is offset, it is assumed that the migration clothing of GPS data is normal distribution. The method of grid initialization based on the probability of GPS locus on the road is put forward, and the corresponding model is established. The two-dimensional median filtering method is used to correct the initial grid data, and the thinning algorithm of lookup table is improved. The improved thinning algorithm is used to refine the raster image of road network to obtain the information of road network skeleton, and the tracking method is used to realize road vectorization. The experimental results show that the coverage rate of this method is 6.43 / 11.54 higher than that of the traditional grid method, and the error rate is reduced by 42.13 and 83.02, which provides an effective method for road network extraction and reveals the important influence of grid size on the result of road network extraction.
【作者单位】：东北大学资源与土木工程学院;
【基金】：国家自然科学基金资助项目(51674063) 国家重点研发计划资助项目(2016YFC0801608)
【分类号】：P228.4;TD67

【相似文献】