HDCRS Summer School 2022
Welcome to the summer school organized by the High-Performance and Disruptive Computing in Remote Sensing (HDCRS) Working Group. HDCRS is part of the IEEE Geoscience and Remote Sensing Society (GRSS), in particular of the Earth Science Informatics (ESI) Technical Committee.
This school is the perfect venue to network with students and young professionals, as well as senior researcher and professors who are world-renowned leaders in the field of remote sensing and work on interdisciplinary research with high performance computing, cloud computing, quantum computing and parallel programming models with specialized hardware technologies.
What the participants said...
"It was very important for me to learn new developed tools and methodology from data collection and recording to applications by large companies like Nvidia, Microsoft, Nasa and many labs!"
"Invaluable new connections with other researchers and engineers working in similar fields to my PhD, rich discussions with them, and many new ideas/insights."
"The coffee breaks and the lunches were amazing and they made this summer school a very smooth experience."
"This summer school has opened my eyes to an interesting set of challenges and given me the motivation to work hard and learn more!"
"Thanks for organizing the school! It was a great experience!"
Lecture topics and instructors
Day 1: Opening
Opening Introduction
Welcome at the University of Iceland and opening of the summer school with an introduction to the IEEE Geoscience and Remote Sensing Society (GRSS).
The Instructors
Prof. Jón Atli Benediktsson
Biography
Jón Atli Benediktsson received the Cand.Sci. degree in electrical engineering from the University of Iceland, Reykjavik, in 1984, and the M.S.E.E. and Ph.D. degrees in electrical engineering from Purdue University, West Lafayette, IN, in 1987 and 1990, respectively. Since July 1, 2015 he is the President and Rector of the University of Iceland. From 2009 to 2015 he was the Pro Rector of Science and Academic Affairs and Professor of Electrical and Computer Engineering at the University of Iceland. His research interests are in remote sensing, biomedical analysis of signals, pattern recognition, image processing, and signal processing, and he has published extensively in those fields. Prof. Benediktsson is a Highly Cited Researcher (Clarivate Analysis, 2018-2020). He was the 2011-2012 President of the IEEE Geoscience and Remote Sensing Society (GRSS) and was on the GRSS AdCom from 2000-2014. He was Editor in Chief of the IEEE Transactions on Geoscience and Remote Sensing (TGRS) from 2003 to 2008 and has served as Associate Editor of TGRS since 1999.
Lecture content
This presentation will introduce the working group “High-performance and Disruptive Computing in Remote Sensing” (HDCRS) of the GRSS Earth Science Informatics Technical Committee (ESI TC). HDCRS is the organizer of this summer school and its main objective is to connect a community of interdisciplinary researchers in remote sensing who are specialized on high-performance and distributed computing, disruptive computing (e.g., quantum computing) and parallel programming models with specialized hardware (e.g., GPUs, FPGAs). The activities of the working group include educational events, special sessions and tutorials at conferences and publication activities, which will be presented.
Get slides
The Instructors
Prof. Dora Blanco Heras
Biography
Dora B. Heras is an associate professor in the Department of Electronics and Computer Engineering at the University of Santiago de Compostela (Spain). She received a MS in Physics in 1993 and was awarded a PhD cum laude from this university. In the period from 2005 to 2010 she was appointed as the head of the Sustainable Development Office at this university. Since 2008 she is also with the research centre CiTIUS (Centro de Investigación en Tecnoloxías Intelixentes) where she leads the hyperspectral remote sensing computing line and has received the accreditation as full professor in 2020. He is also co-chair of the High-Performance and Disruptive Computing in Remote Sensing (HDCRS) Working Group of the IEEE GRSS ESI Technical Committee.
Her research contributions cover a range of topics in the combined fields of image processing, remote sensing, machine learning and high performance computing. In particular, in the last ten years her research has been framed in the line of high performance computing and its application to remote sensing. She has participated in research projects funded by Spanish and European institutions, and R&D agreements. She has served as program committee, guest editor and reviewer in several conferences, in particular, the Euromicro 2021 Parallel and Distributed Conference, and serves as reviewer for different top-ranked journals. She is also a member of the Euro-Par conference Steering Committee since 2018 and has acted as co-chair of the co-located workshops for all the editions since 2017.
Dr. Gabriele Cavallaro
Biography
Gabriele Cavallaro received the B.Sc. and M.Sc. degrees in telecommunications engineering from the University of Trento, Italy, in 2011 and 2013, respectively, and the Ph.D. degree in electrical and computer engineering from the University of Iceland, Iceland, in 2016. He is currently the head of the ‘‘AI and ML for Remote Sensing’’ Simulation and Data Lab at the Jülich Supercomputing Centre, Germany. He is also the chair of the High-Performance and Disruptive Computing in Remote Sensing (HDCRS) Working Group of the IEEE GRSS ESI Technical Committee.
He was the recipient of the IEEE GRSS Third Prize in the Student Paper Competition of the IEEE International Geoscience and Remote Sensing Symposium (IGARSS) 2015 (Milan – Italy). His research interests cover remote sensing data processing with parallel machine learning algorithms that scale on high performance and distributed systems. He serves on the scientific committees of several international conferences and he is a referee for numerous international journals. Since 2019 he gives lectures on scalable machine learning for remote sensing big data at the Institute of Geodesy and Geoinformation, University of Bonn, Germany.
Lecture content
The AI-enhanced Quantum Initiative for EO is a recent initiative from ESA Earth Observation Programmes to assess the potential of Quantum Computing for Earth Observation (QC4EO). Indeed, quantum computing has the potential to improve performance, decrease computational costs and solve previously intractable problems in EO by exploiting quantum phenomena. In this talk we will present our current activities for discovering possible synergies between QC and EO, exploring first promising use-cases, and gathering both communities to prepare the ground for the opportunities which will arise with quantum computing developments.
The Instructors
Dr. Bertrand Le Saux
Biography
Bertrand Le Saux (Ms. Eng. 1999, MSc. 1999 INP Grenoble, PhD 2003 Univ. Versailles / Inria, Dr. Habil. 2019 Univ. Paris-Saclay) is a scientist with the European Space Agency ESRIN/Phi-lab, in Frascati (IT). He is working on data-driven techniques for visual understanding, with a background in machine learning and computer vision. He’s interested in tackling practical problems that arise in Earth observation, to bring solutions to current environmental and societal challenges.
He has been a researcher at CNR/ISTI Pisa (IT), Univ. of Bern (CH), ENS Cachan (FR) and ONERA (FR). He was co-chair [2015-2017] and chair [2017-2019] for the IEEE GRSS technical committee on image analysis and data fusion (IADF TC). He is currently an associate editor of the Geoscience and Remote Sensing Letters. He is also a co-organiser of the CVPR / EarthVision workshop series.
Lecture content
How can we better understand our planet’s past, act on the present issues and predict its future challenges? What if we could generate a digital replica of the Earth as a whole and unveil its inner functioning? This is not a science fiction, but what ESA is currently working on for the Destination Earth project in collaboration with scientists from ECMWF and EUMETSAT: a Digital Twin of the Earth. This talk will present an overview of the Destination Earth project, various related activities as well as the vision for moving forward within this challenging and ambitious endeavor.
Get slides
The Instructors
Dr. Claudia Vitolo
Biography
BiographyClaudia is a senior scientist for the European Space Agency Centre for Earth Observations, where she works on Digital Twin Earth Applications. She previously worked as scientist for Diagnostic and Environmental Applications teams at the European Centre for Medium-range Weather Forecasts and as researcher in various academic institutions. Claudia’s expertise is in geospatial data analysis, web services and artificial intelligence applied to Earth Observations, weather-driven disaster forecasting and other environmental applications.
Claudia lectured in Geographic Information Systems and Data Analysis, maintains a number of open-source projects and she is also an associate editor for the Geoscience Data Journal of the Royal Meteorological Society (Wiley). She co-founded two organisations: the R-Ladies Global and WomenInGeospatial+. Claudia served as voting member of the R-Consortium Infrastructure Steering Committee and was awarded the Microsoft Most Valuable Professional Award.
Lecture content
This lecture will cover geospatial data analysis with the Planetary Computer. We’ll see how the Planetary Computer’s STAC API enables searching a large data catalog to find just the items of interest. We’ll use the Planetary Computer Hub to access data directly from Azure Blob Storage using compute in the same region. We’ll parallelize our computation onto a cluster of machines using Dask and Dask Gateway.
The Instructors
Tom Augspurger
Biography
BiographyTom Augspurger is a software engineer at Microsoft where he works on the Planetary Computer. He contributes to several open-source libraries, including pandas and Dask.
Day 2: Supercomputing and Distributed Computing
Lecture content
In parallel programming we frequently find computing patterns that are common in parallel algorithms. An example is the stencil computation where each output element is computed with data from its neighborhood defined by a mask. For this computation the number of operations per pixel is very large. We will see in this use case how to speed up two common pre-processing steps in hyperspectral analysis using OpenMP, from spectral feature reduction using wavelets to morphological profiles by means of the same parallel pattern.”
Code repository
Get slides
The Instructors
Prof. Pablo Quesada Barriuso
Biography
Pablo Quesada Barriuso is an assistant professor in the Department of Electronics and Computer Engineering at the University of Santiago de Compostela (Spain). He received a MS in Computer Graphics, Video Games and Virtual Reality in 2010 and an awarded cum laude PhD in Information Technology Research in 2015. He is member of the hyperspectral computing group within the department and member of the High-Performance and Disruptive Computing in Remote Sensing (HDCRS) Working Group of the IEEE GRSS ESI Technical Committee.
Pablo has contributed to the state of the art in remote sensing and high-performance computing with several works mainly focused on feature extraction and efficient schemes of segmentation and classification of hyperspectral images which have been published in top-ranked journals and conferences.
His research interests include multi / hyper spectral remote sensing data processing in low-cost devices focused on feature extraction and analysis of data to reduce transfer time, save storage space, and speed-up data analysis for real time processing by programming parallel applications in shared memory architectures and general-purpose computing on GPU.
Lecture content
Image registration is an essential task in many applications of multispectral remote sensing images. Before any processing, the images must be aligned. This process requires a high computational cost that can be alleviated by carrying out implementations for specialized hardware such as GPUs or shared or distributed memory systems.
The Instructors
Dr. Álvaro Ordóñez Iglesias
Biography
Álvaro Ordóñez received the Ph.D. in Computer Science in 2021, Master’s in Big Data Technologies in 2016, and Bachelor’s degree in Computer Science in 2015, all from the Universidade de Santiago de Compostela (USC). Currently, he is postdoctoral researcher at CiTIUS (Centro Singular de Investigación en Tecnoloxías Intelixentes) working on remote sensing image processing. He has collaborated in the organisation of the International European Conference on Parallel and Distributed Computing (Euro-Par) and is a member of the High-Performance and Disruptive Computing in Remote Sensing (HDCRS) Working Group.
His research interests are in pre-processing and registration of multi and hyperspectral images in high performance computing, including NVIDIA GPUs and shared and distributed-memory systems.
Lecture content
Missing from site???
The Instructors
Marcel Aach
Biography
Marcel Aach is a Ph.D. student at the Juelich Supercomputing Centre and the University of Iceland. He obtained his M.Sc. in Economathematics from the University of Cologne in 2021. His research interest include large scale machine learning on high-performance computing (HPC) systems with a focus on distributed hyperparameter optimization and neural architecture search.
Rocco Sedona
Biography
Rocco Sedona received the B.Sc. and M.Sc. degrees in information and communications engineering from the University of Trento in 2016 and 2019, respectively. He is member of the ‘‘High Productivity Data Processing’’ (HPDP) research group at the Jülich Supercomputing Centre, Germany. He is currently pursuing the Ph.D. degree in computational engineering at the University of Iceland.
His research interest is mainly in machine learning methods for remote sensing applications, with a particular focus on distributing deep learning models on multiple GPUs of High Performance Computing (HPC) systems.
Lecture content
In this session, we will cover deep learning and how to achieve scaling to high performance computing systems. We will start the lecture with a presentation of high performance computing system architectures and the design paradigms for HPC software. Furthermore, we give a recap of important machine learning concepts and algorithms. Afterwards, we introduce how deep learning algorithms can be parallelized for supercomputer usage with Horovod. Furthermore, we discuss best practicies and pitfalls in adopting deep learning algorithms on supercomputers and show a practical use case from remote sensing where HPC is used to accelerate training and hyperparameter tuning.
Get slides
Day 3: Specialized Hardware Computing
Lecture content
In this session a general overview of a generic GPU parallel model will be first provided. Then, we will go into the details more closely related to NVIDIA GPU architecture that will be useful for better understanding the key aspects of CUDA programming model. A basic introduction to CUDA programming model will be provided, coupled with some examples specifically related to image processing and remote sensing. These examples will be used as starting point in the second session.
The Instructors
Dr. Raúl Guerra
Biography
Raul Guerra received the Electrical Engineering degree from the university of Las Palmas de Gran Canaria, Spain, in 2012, the master’s degree in telecommunications technologies from the Institute of Applied Microelectronics, and the Ph.D. degree in telecommunications technologies from the University of Las Palmas de Gran Canaria, in 2017. He was funded by the University of Las Palmas de Gran Canaria to do his Ph.D. research in the Integrated System Design Division. In 2016, he was a Researcher with the Configurable Computing Lab, Virginia Tech University. In 2018 he was awarded the prize for the best Ph.D. Thesis in remote sensing finished during 2017 in Spain by the IEEE GRSS. Currently, he works as a postdoctoral researcher in the University of Las Palmas de Gran Canaria where he also teaches image processing and electronics. His research interests include the development of algorithms for images processing and their hardware implementation.
Lecture content
In this session a particular use of NVIDIA LPGPUs for remote sensing applications will be shown. It targets an specific application were a NVIDIA developer kit (Jetson Nano or Jetson Xavier NX) is used as on-board computer within an UAV. On one side, the on-board computer is used to control the flight mission and the hyperspectral data acquisition. On the other side, the on-board computer is also used to compressed the acquired hyperspectral frames in real time. The goal is to first provide a general overview of the application, its requirements and limitations. Then the solution that have been developed will be shown, paying special attention to the CUDA implementation of the compression algorithm, as well as the parallelization strategies used.
The Instructors
Dr. Raúl Guerra
Biography
Raul Guerra received the Electrical Engineering degree from the university of Las Palmas de Gran Canaria, Spain, in 2012, the master’s degree in telecommunications technologies from the Institute of Applied Microelectronics, and the Ph.D. degree in telecommunications technologies from the University of Las Palmas de Gran Canaria, in 2017. He was funded by the University of Las Palmas de Gran Canaria to do his Ph.D. research in the Integrated System Design Division. In 2016, he was a Researcher with the Configurable Computing Lab, Virginia Tech University. In 2018 he was awarded the prize for the best Ph.D. Thesis in remote sensing finished during 2017 in Spain by the IEEE GRSS. Currently, he works as a postdoctoral researcher in the University of Las Palmas de Gran Canaria where he also teaches image processing and electronics. His research interests include the development of algorithms for images processing and their hardware implementation.
Lecture content
This talk will be an introduction to NVIDIA and the vital tools necessary for remote sensing in the Artificial Intelligence (AI) & virtual world era. NVIDIA invests both in internal pure research and platform development to enable a diverse customer base, across gaming, VR, AI, robotics, simulation, digital twinning, graphics, real-time rendering, autonomy, HPC & more. You will be introduced to the hardware and software platform at the heart of this; NVIDIA GPU Computing & will gain insights into how academia, enterprise and startups are applying AI, deploying it in Space as well as designing and testing in readiness. You will also gain a glimpse into state-of-the-art research from world-wide laboratory collaborations & internal work at NVIDIA, demoing, for example, how to illuminate the Moon’s craters. Finally this talk will provide a deeper dive into the workflow from satellite to insight. Beginners might like to try some of our classes using GPU’s in the cloud: www.nvidia.co.uk/dli (codes available on request).
The Instructors
Alison Lowndes
Biography
Alison Lowndes (Senior Scientist, Global AI | NVIDIA). Joining in 2015, Alison spent her first few years as a Deep Learning Solutions Architect and is now responsible for applied Artificial Intelligence both around the globe and off Earth, in Space. A mature graduate in AI, Alison combines technical and theoretical computer science with astrophysics & over 25 years of experience in international project management, entrepreneurial activities and the internet. She consults on a wide range of applications, including planetary defence with NASA, ESA & the United Nations and works closely with world governments advising them on how to harness AI for economic growth, national security & climate action, using NVIDIA’s GPU Computing platform.
Dr. May Casterline
Biography
Dr. May Casterline is a data scientist/image scientist/software developer with a background in satellite and airborne imaging systems. Her research interests include deep learning, hyperspectral and multispectral imaging, innovative applications of machine learning approaches to remote sensing data, multimodal data fusion, data workflow design, high performance computing applications, and creative software solutions to challenging geospatial problems. She holds a PhD and Bachelors of Science in Imaging Science from Rochester Institute of Technology, with a focus on remote sensing. In industry she has acted as a product owner, technical lead, lead developer, and image scientist on both research initiatives and development projects. As a Senior Solutions Architect at NVIDIA, Dr. Casterline works with both industry and government to help enable developers, engineers, data scientists and analysts integrate artificial intelligence and GPU-accelerated solutions into their workflows and products.
Day 4: Quantum computing for Earth Observation
Lecture Content
Principles of quantum computing
- Quantum systems, states, evolution, and measurement
- Quantum computing with ideal and noisy devices
Models of quantum computing
- Gate model of quantum computation
- Quantum annealing and Quadratic Unconstrained Binary Optimization problems
- Elements of quantum machine learning
Quantum computers programming
- Existing quantum hardware
- Programming quantum computers
- Future of quantum computing: risks and opportunities
Review of applications of quantum computing for Earth Observations
- Quantum neural networks
- Quantum supported neural networks
- Combinatoric optimization
Prerequisites
It is required that participants have basic knowledge of linear algebra, probability theory, and machine learning.
The Instructors
Prof. Piotr Gawron
Biography
Piotr Gawron is an applied computer scientist, leader of the Scientific Computing and Information Technology group at AstroCeNT.
He graduated in 2003 from the Faculty Of Automatic Control, Electronics and Computer Science of the Silesian University of Technology, and then obtained PhD in 2008 at the Institute of Theoretical and Applied Informatics, Polish Academy of Sciences where he was employed from 2001 to 2019 in Quantum Systems of Informatics group. His PhD thesis concerned quantum programming languages and their applications for quantum walks and quantum games. In the years 2007—2015 he intensively collaborated on restricted numerical ranges and shadows of matrices with Prof. Karol Życzkowski from the Jagiellonian University. He obtained a habilitation degree from his Alma Mater in 2014 on the basis of the dissertation “Influence of the environment on quantum information processes”. He participated in several research and development projects concerning image processing and energy usage profiling and prediction implemented with industrial partners. He is active in scientific outreach in the area of quantum computation and information. He was a supervisor of two PhD candidates.
The scientific interests of Piotr Gawron concentrate on quantum computation, quantum information theory, machine learning, tensor networks, and recently Earth observations.
Aleksandra Krawiec
Biography
Aleksandra Krawiec got her M.Sc. in mathematics from the Silesian University of Technology. In 2017, she joined the Institute of Theoretical and Applied Informatics, Polish Academy of Sciences, where she is a Ph.D. student in the Quantum Systems of Informatics Group. Her scientific interests focus on quantum information theory, mainly quantum channels’ discrimination and certification. In her free time, Aleksandra enjoys traveling, rock climbing, and dancing folk dances.
Organizers
- Chairs of HDCRS (Gabriele Cavallaro, Dora Blanco Heras, Jin Sun)
- Simulation and Data Lab Remote Sensing (University of Iceland)
In cooperation with and sponsored by
Information overview
- Date and time
Starts 30 May 2022 09:30
Ends 2 Jun 2022 17:00
Greenwich Mean Time (GMT)
- Registration
Start date: 01 January 2022
End date (participate in person): 01 February 2022
End date (access live streaming): 27 May 2022
The school is free of charge
- Contact and Support
Contact person: Dr. Gabriele Cavallaro
Email for questions: hdcrs_chairs@grss-ieee.org
- Venue
Room 132 in Askja (Natural Science Building, University of Iceland), Sturlugata 7, 102 Reykjavík, Iceland. This school will be held in hybrid format.
- Travel Grants
The summer school will welcome up to 30 students on the site. There will be travel grants of up to $1000 for 10 students. The live streaming event will have unlimited capacity.
- Slides and Recordings
The lectures will be recorded and made available online through the GRSS YouTube channel. Course material will be available after the event.
