Model Based Analysis for the Extraction of Forest Stand Information Based on Texture of High-Resolution Optical Remote Sensing Images

Forest ecosystems play a crucial role in the global carbon cycle. Efficient acquisition of forest structural parameters is important for forest inventories and global carbon balance studies. In our research, we simulated the forest canopy scene under different conditions based on the forest optical multi-angle stereo observation model "LandStereo", and used the FOTO texture description method to construct relationships between texture factors and stand density, canopy size and aboveground biomass of the forest. The study explored the role of different spatial resolutions and observation angles on the texture information of images. And we verified the ability to obtain forest stand information based on texture information from real UAV images, in order to qualitatively and quantitatively analyze the potential of FOTO texture factors on estimating the forest stand information.