Zagreb, Croatia, 1 - 5 July 2024

Short Courses in Thermography


In addition to the Main Technical Program, the Conference will be complemented by one-day Short Courses on Monday, 1 July 2024. Short Courses will be held at the International hotel, in a room adjacent to main conference room.

Short Courses Fee: 200,00 EUR (VAT of 25% included)
(does not include participation at the Conference!)


Jean Dumoulin received Ph.D. in Energetic Systems from the National Institute of Applied Sciences (INSA) of Toulouse in 1994 and is specializing in disturbed aerodynamics wall heat transfer identification by infrared thermography (research work conducted in the French National Aerospace Research Institute ONERA). From 1994 to 1997, he was associate research engineer in the former Department of Studies and Research in Mechanics and Energetic Systems at ONERA and associate lecturer at INSA Toulouse. He joined the French Institute of science and technology for transport, development and networks (IFSTTAR: previously LCPC - Laboratoire Central des Ponts et Chaussées) in 1997. Since 2012, he is Associate Professor at Laval University (Quebec, Canada) in the research team MiViM (Canada Research Chair in Multipolar Infrared Vision). In 2013, at its creation, he was invited to join the Inria and IFSTTAR joint research team I4S (Statistical Inference for Structural Health Monitoring). He is currently involved as scientific officer at the European Geosciences Union (EGU) for the Division on Geosciences Instrumentation and Data Systems (GI), and member of the steering committee of the Quantitative Infrared Thermography (QIRT) international scientific organization.


The main topics are:

  • uncooled infrared cameras: main features
  • simplified radiometric heat balance: in-situ environmental corrections approaches
  • apparent surface temperature estimation: toward joint estimation of emissivity and temperature.


Xavier Maldague is a professor at the Department of Electrical and Computing Engineering of the Laval University in Québec (QC), Canada. He has trained more than 50 graduate students (MSc and PhD) and has more than 300 publications. His research interests are in infrared thermography, non-destructive evaluation (NDE) techniques and vision/digital systems for industrial inspection. He holds a Tier 1 Canada Research Chair in Infrared Vision. He is also a chairs of the Quantitative Infrared Thermography (QIRT) Council and a fellow of the Canada Engineering Institute, a Honorary Fellow of the Indian Society of Nondestructive Testing and a fellow of the Alexander von Humbolt Foundation in Germany.


The presentation will deal with the following topics: introduction, theory (radiometry and heat transfer considerations), modelling for 1D, 2D, 3D geometry in solids materials, thermal stimulations in the active approach, infrared detectors and experimental techniques, deployment, data processing and applications.
It is expected this short course will enable the attendees to grasp the fundamentals of IR thermography for non-destructive testing.

Stefano SFARRA

Stefano Sfarra attained a Ph.D. title in mechanical, management and energy engineering at the University of L'Aquila (UNIVAQ), Italy, in 2011. Following the achievement of the Ph.D., he was a research fellow at UNIVAQ until 2017, before becoming a researcher in October of the same year. He carried out research and/or teaching periods abroad at prestigious institutions all over the world. He was also an invited-scientific researcher at Tomsk Polytechnic University (Tomsk, Russia), as well as a member of several scientific committees at international congresses. He is also an editor of Mathematical Problems in Engineering (Hindawi), Infrastructures (MDPI), and Sensors (MDPI). Since December 2020, he is the editor-in-chief of the Quantitative InfraRed Thermography (QIRT) Journal (Taylor & Francis). He is deeply involved in the non-destructive evaluation and characterization of materials, especially using optical and infrared vision non-destructive testing techniques, numerical simulations centered on heat transfer phenomena (by Comsol Multiphysics), development of ad hoc scripts in Matlab environment. In these research areas, he authored - co-authored more than 250 articles in Journals and International Conferences. He also have written six chapters in Books. He is currently acting as a reviewer of 50 scientific journals and collaborator and local contact person in international research projects. He is also a member of the Associazione MASTER and the Associazione Italiana Prove Non-distruttive - Monitoraggio Diagnostica (AIPnD). He received many awards, mainly focused on scientific recognition. In October 2020, he became associate professor at UNIVAQ. Since November 2022, he is an adjunct professor at Laval University (Canada).


Cultural heritage is in danger. Historic houses, medieval walls, castles, antique furniture, movable objects such as panel paintings, paintings on canvas, frescoes, marquetries, are under siege. In fact, severe weather events, invasive plants, no preventive maintenance, earthquakes and pollution are threatening their precious integrity that has stood for centuries. Without expert conservation our heritage could crumble, their stories lost forever, and our landscape/memories be changed irrevocably. The transmission of the knowledge to future generations is pivotal for our society to keep our past alive and move towards the future. The transmission of the knowledge is in the hands of three important figures, i.e., art historians, restorers and scientists. The latter have a key role in the restoration and conservation of cultural heritage objects because, thanks to a clever use of non-destructive testing (NDT) techniques (initially tested on mock-ups), they are able to provide fundamental information to the remaining figures. NDTs help to detect incipient defects before damage becomes visible to the naked eye. Applying a conscientious protocol of diagnostics, in which each object is monitored in the course of time, money can be saved and the “decorative layers” are preserved. Infrared thermography (IRT) method falls in the 'infrared vision' category. IRT, whether appropriately applied in combination with advanced algorithms, is one of the NDTs providing precious information on the health status of artworks. In the presentation, a short review on the use of 'infrared vision' (and in particular of IRT) for the inspection of cultural heritage objects is provided. In particular, a series of experiences done by the author together with esteemed colleagues will be discussed; also, the main contribution of the mock-ups will be stressed.


Gunther Steenackers is a head of Department of Electromechanics at the University of Antwerp and full professor at InViLab Research Group.


About Lecture

Bogusław WIĘCEK

Bogusław Więcek is working in the field of infrared thermography, mainly with respect to its applications to medicine, non-destructive testing, and IR spectroscopy. In addition, his complementary research area is thermal modelling and measurements of electronic and biomedical multilayer, non-homogenous, anisotropic and non-linear structures. He was a chief of the research group developing the first in Poland metrological microbolometer camera based on VOx detector. Currently, he is working on photonic system applying the Raman scattering in NIR for temperature and material content measurements using low-power lasers. He was a supervisor of 9 Ph.D. dissertations. He is a co-author of 13 patents. He is the chairman of the International Conference on Infrared Thermography and Thermometry. Currently, he serves on the editorial board of three journals and scientific committees in six organizations.


The lecture will mainly focus on Deep Learning systems for solving regression and classification problems in infrared signal and image processing. The different architectures of Convolutional Neural Networks (CNN) will be discussed, including 2D filters and feature layers, connections between layers, activation functions, feature map size reduction (pooling), padding, normalization, etc.

Loss functions and typical learning algorithms will be discussed. In the learning phase, elements of knowledge acquisition will be presented underlying the transfer learning, enriching training datasets (augmentation), regularization methods and preventing overfitting. The problem of gradient vanishing while training with some optimization methods and the use of ResNet networks with bypass connections will be highlighted.

An example implementation of a CNN network in the popular and frequently used TensorFlow/Python environment using the Keras library will be demonstrated. Available pre-trained CNNs will be presented, such as AlexNet, GoogleNet, VGGNet and other "light" networks for remote applications.