These models were created using the
DeepForest python library. The library uses Pytorch but simplifies the interface for non-ML experts.
DeepForest provides a pre-trained convolutional neural network model capable of detecting tree crowns from aerial imagery.
However, it may not work perfectly in your area of interest. It may produce false positives (ie, objects that are not actually trees) and false negatives (ie, fail to identify all trees in an image). To improve the pre-trained model, users can incorporate their own training data to fine-tune the model to identify your specific trees of interest.
DeepForest uses RetinaNet, which is one stage object detection model. RetinaNet consists of two major components, 1. a pre-trained convolutional neural network called ResNet50 and 2. a feature pyramid network (FPN). For a deeper look at the model architecture, please see
this article.
