People from Ghent University and Van Eyck: engineering

(02-02-2020) Ghent’s 2020 focus is on Van Eyck and the Ghent Altarpiece. Lots of people from Ghent University are involved in this year of celebration. Over the coming weeks we put them in the spotlight. This week: Aleksandra Pizurica and Arnold Janssens

In Ghent, it’s all about Van Eyck and the Ghent Altarpiece in 2020. One of the highlights is undoubtedly the return of the restored painting to St. Bavo’s cathedral. Years of research preceded the amazing renovation. Moreover, many people from Ghent University have contributed with new insights relating to the Ghent Altarpiece. You have already gained a glimpse into these insights over the past few weeks. Today part 3: engineering.

Professor Aleksandra Pizurica (department of Telecommunication and Information Processing, Group for Artificial Intelligence and Sparse Modelling - GAIM) used AI to reveal what had always been hidden. Plus, to help ensure that there was even a masterpiece left to restore, professor Arnold Janssens (department of Architecture and Urban Planning) researched how to improve the indoor climate in St. Bavo’s cathedral. 

What is the link between your area of expertise and research into the Ghent Altarpiece? 

Aleksandra Pizurica: “Ever since 2010, our research group has been trying to use digital imaging and machine learning - our expertise at GAIM - to answer questions relating to art history and to support (preparations for) restauration. We have used our technology in two areas: to analyse the painting style of Jan Van Eyck and to support the restauration and conservation via virtual restauration. Our work is cutting edge in both areas.”

Arnold Janssens: “One of our research directions concerns improving the indoor museum climate in order to protect works of art, and developing numeric models to do so. As a result of our research, the restauration team contacted us over a decade ago, and asked us to analyse the climate issues in St. Bavo’s cathedral and consider how they could be resolved. This concerned matters such as torn canvas and flaking paint.”

What insights were gained as a result of your research?

Aleksandra Pizurica: “By analysing the many painted pearls and gemstones, we were able to characterise the ‘signature’ in painting style. This allowed us to express objectively and quantitatively differences between the painting methods of Jan Van Eyck, his copyists and other painters. The consistency of the painterly execution (e.g. the position of highlights and the overall impression of the interaction with the light source) also varied depending on where the figures were positioned on the panels. We assume that Jan executed with most attention the pearls in the prominent places of the altarpiece, and might have entrusted others in his atelier to take care of the smallest and less visible ones. This can also indicate possible adaptations, assisting this way the restauration.”


“The second part of our research involved studying paint loss and cracks in the layers of varnish and in different paint layers, that are often invisible to the naked eye. In this way, we have managed to improve the readability of the inscriptions. For example, in the book on the panel about the Annunciation that had always remained a mystery, our virtual crack inpainting facilitates the deciphering. We have also developed a tool to detect paint loss. I believe that Ghent University has been a pioneer in this respect: we were the first to develop automated paint-loss detection and the first ones to employ deep-learning for crack detection in paintings. We are now turning these models into handy tools, suitable for use by those carrying out the restauration.”


Arnold Janssens: “After analysing the indoor climate we offered advice on the most urgent measures. Due to the lack of heating in the building, the relative humidity reached high levels in the winter, risking organic damage. The fluctuations in climate were mainly linked to the position of the chapel, right beside the entrance to the cathedral, where the doors were kept permanently open. This posed the risk of panel distortion and cracks in the paint.”


“Therefore, we recommended installing heating and a porch at the entrance to the cathedral. The church followed both sets of advice. The shrine was equipped with an on-site humidifier as protection from extremely dry conditions. Also, the large batches of visitors were causing risky short-term fluctuations, with their damp coats, respiration, body heat, etc. So we suggested managing visitors in smaller groups. However, this has never been implemented.”


“Following these procedures, we took new measurements. This led to new initiatives, such as training for curators and the installation of a permanent monitoring system.”

Who from Ghent University was involved in this research?

Aleksandra Pizurica: “Our research group worked closest of all with Ghent University’s professor Maximiliaan Martens, professor-emeritus Mark De Mey and doctoral student Hélène Dubois, as the restauration coordinator. Several doctoral students from our group were involved: Ljiljana Platisa (now postdoc and business developer, i-Know consortium), Tijana Ruzic (now at Televic), Laurens Meeus, Roman Sizyakin and Shaoguang Huang.”


Arnold Janssens: “Above all, doctoral students Lien De Backer and Marijke Steeman – who is now an assistant professor in the department - and professor Michel De Paepe (Department of Electromechanical, Systems and Metal Engineering). Michel joined me in promoting the various research projects relating to the indoor museum climate.”

Now that the Ghent Altarpiece has been returned to St. Bavo’s cathedral you can go and discover it for yourself. How many pearls can you count in the painting? Can you decipher the inscriptions with your naked eye? Are there not too many visitors crowding around the mystic lamb? Oh yes, and be sure to take your time to enjoy this masterpiece! However, don’t forget to wear your warm coat, as the lamb can’t stand the heat.

Next week: archaeology, more specifically: archeometry