Semantic Segmentation of Remote sensing image

Abstract

This dissertation’s objective is to gain a thorough understanding of the field of image semantic segmentation as well as a deep dive into the field of deep learning and how we can use different networks, particularly convolutional neural networks and, more specifically, the application of VGG, Resnet, Inception, and U-Net architectures in the Remote Sensing imagery. The experiments and evaluations of the aforementioned models show the benefits of using such networks as well as their efficiency in resolving several problems with image semantic segmentation.