CNN model to map vegetation classification in a subalpine coniferous forest using UAV imagery

This study reveals the potential of combining deep learning with UAV remote sensing for natural forests. This approach can help to obtain forest information more reliably in inaccessible forest areas, providing an efficient and cost-effective method for monitoring and protecting natural forests. The code was written using the interactive programming environment Jupter notebook and developed using the arcgis.learn module in the ArcGIS API for Python (https://developers.arcgis.com/python/api-reference/). The code is easier to use in Jupter notebook, which is built into ArcGIS Pro. The dataset includes the drone image and the sample set. The UAV image was captured using DJI Mavic 2 (SZ DJI Technology Co., Shenzhen, China). The sample set includes a training set and a test set, which can be tested by building models using the code in the Code folder.