Gobi Vegetation Recognition Based on Low-Altitude Photogrammetry Images of UAV

2018 International Conference of Green Buildings and Environmental Management

土木建筑工程;生态环境科学

Yang, Demeng^1

College of Liberal Arts and Sciences, National University of Defense Technology, Changsha

410073, China^1

关键词: Multiscale segmentation;    Near infrared band;    NorthWest China;    Remote sensing images;    Spectral characteristics;    Technology-based;    Vegetation distribution;    Vegetation index;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/186/5/012053/pdf
DOI  :  10.1088/1755-1315/186/5/012053

来源: IOP

【 摘 要 】

Vegetation distribution information is crucial to the environmental protection of Gobi. The traditional method to obtain vegetation distribution information is building vegetation index which can be used as the threshold for vegetation recognition, with the spectral characteristics of vegetation in remote sensing images form satellite or aviation, especially the near infrared band. In recent years, low-altitude photogrammetry technology based on UAV has developed rapidly. But the UAV images do not contain near-infrared band information, many researchers have adopted visible light spectral characteristics to construct new vegetation index and achieved many progress. In this paper, taking the Gobi region of northwest China as a example, we introduced the methods for multi-scale segmentation and classification of DOM which generated from the UAV images. Further, based on eCognition Developer, we extracted visible light-based vegetation indices including NGRDI, NGBDI, VDVI, GLI, and ExG, and assessed the vegetation recognition effect of these five indices.