Yichang mountainous area tree species identification method based on time sequence remote sensing image and DTW
A Yichang mountainous area tree species identification method based on a time sequence remote sensing image and DTW comprises the following steps: S1, collecting spectral information of trees, and storing the spectral information to obtain a spectral information image; s2, analyzing the spectral inf...
Saved in:
Main Authors | , , , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
07.08.2020
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | A Yichang mountainous area tree species identification method based on a time sequence remote sensing image and DTW comprises the following steps: S1, collecting spectral information of trees, and storing the spectral information to obtain a spectral information image; s2, analyzing the spectral information image, and constructing an NDVI time sequence of time and surface temperature; s3, calculating the DTW distance of the NDVI time sequence, and obtaining a threshold value used for distinguishing Yichang mountain area tree species; and S4, if the NDVI time sequence is greater than the threshold in the step S3, determining that the tree is arbor, otherwise, determining that the tree is shrub. In the present invention, an NDVI time sequence is established through construction time and surface temperature, then the DTW distance of the NDVI time sequence is calculated through a hidden Markov model (HMM) method, a threshold value used for distinguishing Yichang mountainous area tree species is obtained, if the D |
---|---|
Bibliography: | Application Number: CN202010189413 |