Disputas - Master of Science Luigi Tommaso Luppino

Master of Science Luigi Tommaso Luppino will Tuesday June 16 at 12.15 publically defend his thesis for PhD degree in Science.

Title of the thesis:

«Unsupervised Change Detection in Heterogeneous Remote Sensing Imagery»

Popular scientific abstract:

Comparing apples and oranges - and making sense of it

Change detection is an important application in earth observation with satellite images. Change detections can for instance be used to detect deforestation or flooded areas. It is traditionally done by comparing images recorded under conditions that should be as similar as possible. The normal approach is to compare images from the same instrument and sensor mode, such that the measurements at two separate times can be directly compares and where a difference close to zero represents "no change". This requires, however, exact corrections and co-calibration of the images, which is difficult due to common cariation of sensor geometry and environmental parameters that change with weather and season.

The quest posed here is less obvious: How to compare different images from different sensors in order to find change between them? - It is a bit like comparing apples and oranges - in a sensible fashion. The motication is both to perform faster and more extensive analyses. With heterogeneous change detection there is no need to wait for an image taken by the same sensor after an event of interest. Sine the method uses all kinds of images, changes can be detected as soon as any given image is available. The ability to compare different sensors also provides longer and denser time series of data, that can be used to extend the analysis in time span and temporal resolution.

This makes for better exploitation of the vast amounts of satellite images that are captured.

The concept of heterogeneous or multimodal change detection is very new, and the challenge is hereby addressed with advanced methods from statistics and machine learning. The developed methods use the last innovations within artificial intelligence and deep learning, demonstrating their great usefulness for the detection of e.g. forest fires and flooding on time series of heterogeneous images.

Supervisors:

Associate professor Stian Normann Anfinsen, Department of Physics and Technology, UiT (main supervisor)
Professor Robert Jenssen, Department of Physics and Technology, UiT
Professor Sebastiano Serpico, University of Genova
Associate professor Gabriele Moser, University of Genova

Evaluation committee:

Dr. Francesca Bovolo, Remote Sensing for Digital Earth, Fondazione Bruno Kessler, Italy (1. opponent)
Professor Yann Gousseau, Image, Data and Signal Department, Télécom ParisTech, France (2. opponent)
Associate professor Anthony P. Doulgeris, Department of Physics and Technology, UiT (internal member and leader of the committee)

Opposition ex auditorio:

If you have any questions for the candidate during the public defence, please send an e-mail to: camilla.brekke@uit.no. She will announce the questions during the defence.

Leader of the public defence:

Professor Camilla Brekke, Vice Dean of Research, Faculty of Science and Technology, UiT.

Trial lecture:

The trial lecture is held Tuesday June 16 at 10.15 in the same auditorium. The trial lecture will be streamed. Title of the trial lecture:

Audience:

UiT follows the national guidelines regarding infection control. A maximum of 50 people are allowed in the auditorium during the defence, as long as everybody keeps a distance of 1 meter at all times.

The public defence will be streamed from this website.

You can find the thesis here.