论文部分内容阅读
在机载LiDAR(Light Detection and Ranging)数据和高空间分辨率航空影像的支持下,以城市为实验区,实现了单木树冠提取。首先通过LiDAR数据获取高差模型,将其作为包含林木的感兴趣区,再通过掩膜方式提取高分影像上的相同区域,然后采用标记分水岭分割算法分别对两幅感兴趣区影像进行树冠提取,最后以人工勾绘树冠结果为参考评价分割精度,比较了两种数据源提取树冠的优缺点。结果显示,利用LiDAR数据获取的高差模型中包含85.28%的林木信息,林木区域提取的效果显著;基于高分影像得到的分割结果较好,F值为57.14%,基于高度差值模型影像的分割结果较差,F值为42.47%。表明分水岭算法方便可行,且高分影像提供的二维信息更适用于树冠提取。
With the support of airborne LiDAR (Light Detection and Ranging) data and high spatial resolution aerial imagery, the city was taken as the experimental area to realize the single-tree canopy extraction. Firstly, the height difference model is obtained by LiDAR data, which is taken as the region of interest which includes the forest tree, and the same region on the high resolution image is extracted through the mask method. Then, the watershed segmentation is used to extract the canopy images of two regions of interest Finally, the accuracy of the segmentation was evaluated by using the results of artificial crown extraction, and the advantages and disadvantages of two data sources were compared. The results showed that the height difference model obtained with LiDAR data contained 85.28% of the forest information, and the tree area extraction had a significant effect. The segmentation result based on the high-resolution image was better, with an F value of 57.14%. Based on the height difference model image Segmentation results are poor, F value of 42.47%. It shows that watershed algorithm is convenient and feasible, and the two-dimensional information provided by high-resolution image is more suitable for canopy extraction.