使用Deeplabv3+（骨干Resnet50）对无人机图像进行语义分割。.zip

# UAV-Semantic-Segmentation using Deeplabv3+ (backbone Resnet50) Use Deeplabv3+ (backbone Resnet50) to perform Semantic Segmentation for UAV images. ## Usage 1. Run train_pred_aug.m to perform augmentations to images and masks. 2. Run partitionData.m to split traning set and test set. 3. Run hurrican_harvel.m to train the model and make prediction. Remeber to load Deeplabv3+ network layergraph if you don't have it.