A Novel Desert Vegetation Extraction and Shadow Separation Method Based on Visible Light Images from Unmanned Aerial Vehicles

Yuefeng Lu, Zhenqi Song, Yuqing Li, Zhichao An, Lan Zhao, Guosheng Zan and Miao Lu ()
Additional contact information
Yuefeng Lu: School of Civil and Architectural Engineering, Shandong University of Technology, Zibo 255049, China
Zhenqi Song: School of Civil and Architectural Engineering, Shandong University of Technology, Zibo 255049, China
Yuqing Li: School of Civil and Architectural Engineering, Shandong University of Technology, Zibo 255049, China
Zhichao An: School of Civil and Architectural Engineering, Shandong University of Technology, Zibo 255049, China
Lan Zhao: School of Civil and Architectural Engineering, Shandong University of Technology, Zibo 255049, China
Guosheng Zan: Academy of Forestry Inventory and Planning, National Forestry and Grassland Administration, Beijing 100714, China
Miao Lu: Key Laboratory of Agricultural Remote Sensing, Ministry of Agriculture and Rural Affairs, Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 100081, China

Sustainability, 2023, vol. 15, issue 4, 1-20

Abstract: Owing to factors such as climate change and human activities, ecological and environmental problems of land desertification have emerged in many regions around the world, among which the problem of land desertification in northwestern China is particularly serious. To grasp the trend of land desertification and the degree of natural vegetation degradation in northwest China is a basic prerequisite for managing the fragile ecological environment there. Visible light remote sensing images taken by a UAV can monitor the vegetation cover in desert areas on a large scale and with high time efficiency. However, as there are many low shrubs in desert areas, the shadows cast by them are darker, and the traditional RGB color-space-based vegetation index is affected by the shadow texture when extracting vegetation, so it is difficult to achieve high accuracy. For this reason, this paper proposes the Lab color-space-based vegetation index L2AVI (L-a-a vegetation index) to solve this problem. The EXG (excess green index), NGRDI (normalized green-red difference index), VDVI (visible band difference vegetation index), MGRVI (modified green-red vegetation index), and RGBVI (red-green-blue vegetation index) constructed based on RGB color space were used as control experiments in the three selected study areas. The results show that, although the extraction accuracies of the vegetation indices constructed based on RGB color space all reach more than 70%, these vegetation indices are all affected by the shadow texture to different degrees, and there are many problems of misdetection and omission. However, the accuracy of the L2AVI index can reach 99.20%, 99.73%, and 99.69%, respectively, avoiding the problem of omission due to vegetation shading and having a high extraction accuracy. Therefore, the L2AVI index can provide technical support and a decision basis for the protection and control of land desertification in northwest China.

Keywords: land desertification; UAV visible remote sensing imagery; vegetation index; shadow texture; color space (search for similar items in EconPapers)
JEL-codes: O13 Q Q0 Q2 Q3 Q5 Q56 (search for similar items in EconPapers)
Date: 2023
References: View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
https://www.mdpi.com/2071-1050/15/4/2954/pdf (application/pdf)
https://www.mdpi.com/2071-1050/15/4/2954/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jsusta:v:15:y:2023:i:4:p:2954-:d:1059659

Access Statistics for this article

Sustainability is currently edited by Ms. Alexandra Wu

More articles in Sustainability from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().