Abstract This paper introduces an effective semantic segmentation of satellite imagery using 3D-Unet. We discussed previous work carried out and used deep learning to reproduce the work of Kemker et al. (2018) and other methods. We used performance metric to compare the performance of the proposed. This work underline that most RS and DL segmentation can be enhanced using DL models.
Keywords: Computer Vision, Deep Learning, Semantic Segmentation, 3D-Unet, Satellite Imagery.
| DOI: 10.17148/IJARCCE.2021.101207