Building and Road Segmentation Using EffUNet and Transfer Learning Approach

TL;DR

This paper introduces a novel segmentation architecture combining EfficientNetV2 with UNet for urban object detection in satellite imagery, achieving high accuracy on building and road datasets.

Contribution

It proposes a new EfficientNetV2-UNet architecture for improved building and road segmentation from aerial images, leveraging transfer learning.

Findings

Achieved mIOU of 0.8365 for buildings

Achieved mIOU of 0.9153 for roads

Demonstrated state-of-the-art performance on benchmark dataset

Abstract

In city, information about urban objects such as water supply, railway lines, power lines, buildings, roads, etc., is necessary for city planning. In particular, information about the spread of these objects, locations and capacity is needed for the policymakers to make impactful decisions. This thesis aims to segment the building and roads from the aerial image captured by the satellites and UAVs. Many different architectures have been proposed for the semantic segmentation task and UNet being one of them. In this thesis, we propose a novel architecture based on Google's newly proposed EfficientNetV2 as an encoder for feature extraction with UNet decoder for constructing the segmentation map. Using this approach we achieved a benchmark score for the Massachusetts Building and Road dataset with an mIOU of 0.8365 and 0.9153 respectively.