Deep Learning for Understanding Satellite Imagery: An Experimental Survey.

Sharada Prasanna Mohanty,Akshat Tandon,Jan Philip Göpfert,Piotr Tarasiewicz,Janick Rohrbach,Kamil A Kaczmarek,Andrzej Pyskir,Malte Schilling,Marcel Salathe,Dave Luo,Saket Kunwar,Sascha Fleer,Manjunath Prasad,Nikhil Rayaprolu,Guillaume Mollard,Jakub Czakon

doi:10.3389/frai.2020.534696

Abstract

Translating satellite imagery into maps requires intensive effort and time, especially leading to inaccurate maps of the affected regions during disaster and conflict. The combination of availability of recent datasets and advances in computer vision made through deep learning paved the way toward automated satellite image translation. To facilitate research in this direction, we introduce the Satellite Imagery Competition using a modified SpaceNet dataset. Participants had to come up with different segmentation models to detect positions of buildings on satellite images. In this work, we present five approaches based on improvements of U-Net and Mask R-Convolutional Neuronal Networks models, coupled with unique training adaptations using boosting algorithms, morphological filter, Conditional Random Fields and custom losses. The good results—as high as and —from these models demonstrate the feasibility of Deep Learning in automated satellite image annotation.

Highlights

Despite substantial advances in global human well-being, the world continues to experience humanitarian crizes and natural disasters
We focus on the problem of instance segmentation on a simplified version of the SpaceNet dataset, in order to detect buildings in different urban settings on high resolution satellite imagery
We explore different flavors of U-Net and Mask R-Convolutional Neuronal Networks (CNN) on a task of instance segmentation on high resolution satellite imagery to detect buildings

Summary

Introduction

Despite substantial advances in global human well-being, the world continues to experience humanitarian crizes and natural disasters. Long-term and reignited conflicts affect people in many parts of the world, but often, accurate maps of the affected regions either do not exist or are outdated by disaster or conflict. Satellite imagery is readily available to humanitarian organizations, but translating images into maps is an intensive effort. Applications of the state-of-the-art results in deep learning have been increasingly accessible to various different domains over the last few years (LeCun et al, 2015), the main reasons being the advent of end-to-end approaches in deep learning (LeCun et al, 2015), and the access to vast amounts of openly available data and high performance compute. While access to high-performance compute infrastructure has not been an inhibiting factor, access to high-resolution

Objectives

Results

Conclusion