Earth observation

This innovation area aims to develop deep learning models and applications for monitoring and prediction of objects, hazard risks and streamlining of aerial surveys.

Motivation

Optical images from drones or satellites and data captured by radar sensors from above contain enormous amounts of complex data. They have the potential to reveal valuable information about our planet and its surface that could be used automate terrain mapping or to predict objects and hazard risks such as vessels and potential oil spills at sea.

Our innovations

Visual Intelligence researchers have developed novel methods for improving the monitoring and prediction of hazard risks, object detection, and for surveying and mapping the ground and sea from the air by exploiting remote sensing images. Such methods include new approaches for:

  • detecting objects in oblique aerial imagery, in collaboration with former user partner Field.
  • detecting vessels and other objects.
  • detecting oil spills and characterizing the thickness of such spills.

The latter two methods were developed in close collaboration with user partner Kongsberg Satellite Services (KSAT).

Addressing research challenges

Limited and inadequate training data is a general problem in remote sensing. Combinations of multi-sensor data, e.g. from optical and radar sensors, and time dependencies is another key challenge. Modelling of contextual information may also enhance the performance. The methods mentioned earlier address these research challenges in different ways.

For instance, the proposed oil spill detection method is a step towards achieving uncertainty quantification in deep learning models for remote sensing data analysis.

Synergies within the innovation area and across innovation areas

As for all image analysis applications, the development of deep learning methodology to solve certain tasks in earth observation often benefits from solutions developed to solve other problems.For instance, when developing methods for building segmentation, ideas from segmentation of oil spills can potentially be transferred.  

It can also be valuable to reveal any different behaviour of segmentation algorithms due to different properties in data sources. Aerial imagery and satellite imagery come with different resolution, contrast, noise properties, and by contrasting seemingly similar deep learning methods, the influence of different data properties may be revealed and better understood.

We have developed new methods and obtained insights into how one can use self-supervised learning when there are images acquired at two or more times available, both as a pre-training step and as an integral part of change detection. The insights and experience gained here on the concept of self-supervised learning in general has a wide range of other relevant applications. We have explored similar methodology within seismic analysis, where it has been utilized to identify and characterize geological regions in vast seismic datasets.

Highlighted publications

Interrogating Sea Ice Predictability With Gradients

February 14, 2024
By
Joakimsen, H. L., Martinsen I., Luppino, L. T., McDonald, A., Hosking, S., and Jenssen, R.

Multi-modal land cover mapping of remote sensing images using pyramid attention and gated fusion networks

July 1, 2022
By
Qinghui Liu, Michael Kampffmeyer, Robert Jenssen, Arnt-Børre Salberg

Other publications

Better, Not Just More: Data-centric machine learning for Earth observation

By authors:

Roscher, Ribana; Russwurm, Marc; Gevaert, Caroline; Kampffmeyer, Michael Christian; Santos, Jefersson A. Dos; Vakalopoulou, Maria; Hansch, Ronny; Hansen, Stine; Nogueira, Keiller; Prexl, Jonathan; Tuia, Devis

Published in:

EEE Geoscience and Remote Sensing Magazine 2024 s. 1-22

on

October 31, 2024

Interrogating Sea Ice Predictability With Gradients

By authors:

Joakimsen, H. L., Martinsen I., Luppino, L. T., McDonald, A., Hosking, S., and Jenssen, R.

Published in:

IEEE Geoscience and Remote Sensing Letters

on

February 14, 2024

A self-supervised inspired object scoring system for building change detection

By authors:

Jensen, Are Charles

Published in:

Proceedings of Machine Learning Research (PMLR) ISSN 2640-3498. 233, p. 97–103

on

January 8, 2024

Using Deep Learning Methods for Segmenting Polar Mesospheric Summer Echoes

By authors:

Domben, Erik Seip; Sharma, Puneet; Mann, Ingrid

Published in:

Remote Sensing 2023 ;Volum 15.(17) Suppl. 4291. s.1-23

on

August 31, 2023

SAR and Passive Microwave Fusion Scheme: A Test Case on Sentinel-1/AMSR-2 for Sea Ice Classification

By authors:

Khachatrian, Eduard; Dierking, Wolfgang; Chlaily, Saloua; Eltoft, Torbjørn; Dinessen, Frode; Hughes, Nick; Marinoni, Andrea.

Published in:

Geophysical Research Letters 2023 ;Volum 50.(4) s.1-7

on

February 14, 2023