| Urban greening trees are an important part of the urban ecosystem.Trees can purify the air,prevent dust and reduce dust.It can also adjust the microclimate,temperature and humidity in a small area of the city,which plays an irreplaceable role in establishing and maintaining a comfortable and livable urban living environment.Leaf is an important part of trees,and it is also an important organ for tree to obtain nutrients.The spectral characteristics of leaves largely reflect the growth and health status of the trees,which can be used as the most intuitive observation basis for judging the state of the tree.Obtaining the detailed characteristics of leaves is the key to understanding the state of trees and protecting urban greening trees,and it is also a difficult point in the study.The detailed characteristics of leaves include leaf color changes,leaf vein differences and leaf thickness.Spectral characteristics can be used to reflect the differences caused by different leaf details.The accurate extraction of spectral characteristics is the key to studying urban greening trees.In order to extract the spectral characteristics of urban tree leaves,using hyperspectral LiDAR to obtain the spectral information of sample leaves at different measurement positions,and studying the difference of spectral fine characteristics.Based on the effective extraction of the spectral fine characteristics,the research on the effect of dust retention on the spectral characteristics of leaves was carried out.The main contributions are as follows:(1)The spectral data of six broad-leaved tree leaf samples(Fraxinus pennsylvanica,Koelreuteria paniculata,Lonicera japonica,Amygdalus davidiana,Populus tomentosa,Magnolia denudate.)were collected using the hyperspectral LiDAR system built in the laboratory,and collected spectral data of eight different measurement points in each sample.In order to explore the difference of spectral characteristics of different measurement points,a divergence analysis method was proposed to extract the fine spectral characteristics of leaves based on the analysis of leaf spectral characteristics,quantitatively analyzed the spectral fine characteristics of leaves,and analyzed the reasons for the spectral differences.The results showed that the divergence analysis method is feasible for extracting the fine features of leaf spectrum.(2)Dust-retaining leaf samples(Fatsia japonica,Photinia stenophylla,Deyeuxia langsdorffii and Magnolia grandiflora)were collected in specific urban areas,using hyperspectral LiDAR to complete data collection before and after dust removal of the samples in a laboratory environment.The collection process was as follows: First,the spectrum data of the dust-retaining blades was collected.Then,the dust was removed at the collection location.Finally,the spectrum data of the dust-removed blades was collected.Based on the analysis of leaf spectral data before and after dust removal blades,a method for extracting dust-retaining leaf spectral characteristics based on sensitive response ratio was proposed to quantitatively analyze vegetation index(leaf chlorophyll index,leaf water index,ratio vegetation index,red edge index and simple ratio index)to the degree of response to dust retention.Finally,a linear correlation fitting model with dust-retaining leaf vegetation index and sensitive response ratio was established and tested.The results showed that the leaf water index is the least sensitive to dust retention,the ratio vegetation index has the largest sensitivity response ratio,and the leaf chlorophyll index has the highest correlation with the sensitivity response ratio.Figure [18] table [15] reference [60]... |
PDF Full Text Request