Segmentation of Satellite Imagery using U-Net Models for Land Cover Classification

TL;DR

This paper presents a modified U-Net model for satellite image segmentation to improve land cover classification accuracy and change detection, validated on large European datasets with promising results.

Contribution

The study introduces a novel U-Net based model tailored for satellite imagery, demonstrating high accuracy in land cover classification and addressing noisy labels in large datasets.

Findings

High F1 score of 0.749 for multiclass classification

Effective segmentation of forests, waters, and arable land

Potential to enhance land cover maps and change detection

Abstract

The focus of this paper is using a convolutional machine learning model with a modified U-Net structure for creating land cover classification mapping based on satellite imagery. The aim of the research is to train and test convolutional models for automatic land cover mapping and to assess their usability in increasing land cover mapping accuracy and change detection. To solve these tasks, authors prepared a dataset and trained machine learning models for land cover classification and semantic segmentation from satellite images. The results were analysed on three different land classification levels. BigEarthNet satellite image archive was selected for the research as one of two main datasets. This novel and recent dataset was published in 2019 and includes Sentinel-2 satellite photos from 10 European countries made in 2017 and 2018. As a second dataset the authors composed an original set containing a Sentinel-2 image and a CORINE land cover map of Estonia. The developed classification model shows a high overall F\textsubscript{1} score of 0.749 on multiclass land cover classification with 43 possible image labels. The model also highlights noisy data in the BigEarthNet dataset, where images seem to have incorrect labels. The segmentation models offer a solution for generating automatic land cover mappings based on Sentinel-2 satellite images and show a high IoU score for land cover classes such as forests, inland waters and arable land. The models show a capability of increasing the accuracy of existing land classification maps and in land cover change detection.