Analysis of the importance of five new spectral indices from WorldView-2 in tree species classification

作者：Liu, Huaipeng^*; An, Huijun

来源：Journal of Spatial Science, 2020, 65(3): 455-466.

DOI：10.1080/14498596.2018.1521754

• WorldView-2
• tree species classification
• spectral index mining
• importance measure
• MLC-RFI

Spectral information mining plays a signi?cant role in tree species classi?cation using remote sensing images but a measure of the importance of the mined features is also necessary. We created five new spectral indices and used MLC-RFI to determine their importance. The results showed that these spectral indices had a positive effect on tree species classi?cation, and their importance was ranked as follows: PAI(578)> PAI(678)> TAI(578)> TAI(678)> SI57. The experimental results showed that important spectral indices for tree species classi?cation can be created and that it is bene?cial to mine spectral information from remote sensing images.

• 出版日期2020-9-1
• 单位内蒙古农业大学; 洛阳师范学院