Personale docente

Stefano Cagnoni graduated in Electronic Engineering at the University of Florence in 1988. He was a Ph.D. student and a post-doc until 1997, working in the Bioengineering Lab of the Department of Electronic Engineering. He received the Ph.D. degree in Bioengineering in 1993 from the same institution.

In 1994, he was a visiting scientist at the Whitaker College Biomedical Imaging and Computation Laboratory at the Massachusetts Institute of Technology. He held a post-doc position at the University of Florence in 1995 and 1996. Since 1997, he has been with the Department of Engineering and Architecture (formerly Department of Information Engineering) at the University of Parma, where he has served as an Associate Professor since 2004.

Regarding basic research, his primary interests are in machine learning and soft computing, with a particular focus on evolutionary computation. Regarding applied research, his primary areas of interest are the application of such techniques to problems in pattern recognition, classification, and computer vision.

During his early career, he developed Evolutionary Computation (EC) applications, including genetic algorithms for designing filters to analyze one-dimensional signals and for contour-based image segmentation, as well as symbolic (expert systems) and subsymbolic (neural networks) applications to medical data and images.

In the field of evolutionary computation, besides actively participating in EvoNet, the EU-funded network of excellence in from 1997 to 2005, when the project concluded, he developed an early example of a human-in-the-loop application of Genetic Programming (GP) to create false-color images that synthesize information from sequences of grey-level images, using an interactive evolutionary method that uses the user's preferences as fitness criteria. He also used GP to define the optimal play strategy for a robot goalkeeper, which successfully participated in RoboCup 99. Using an original approach based on cellular programming, he also developed a license plate character recognition system, for which he studied efficient GP algorithms resulting in ensembles of extremely efficient binary classification programs. Using such techniques, he could obtain low-resolution character classifiers that showed a tenfold increase in performance, while maintaining nearly the same accuracy as a reference neural classifier.

His EC research has then focused on the theory and applications of Particle Swarm Optimization (PSO). In particular, regarding applications, he has studied computer vision algorithms that utilize PSO to detect and track objects/people. He has also developed some of the first GPU-based implementations of PSO, analyzing the efficiency of different synchronous and asynchronous PSO versions based on graphics hardware. Regarding computer vision, his research primarily focused on low-level techniques, particularly contour-based and region-based image segmentation. In the first phase, it aimed to extract three-dimensional information from sequences of tomographic images, and subsequently, to locate free space in applications of autonomous navigation for robotics. He has used massively parallel computer architectures based on cellular automata to implement algorithms for perspective effect removal and to compute the optical flow field. He also designed a hybrid sensor prototype that combines a traditional camera with an omnidirectional one. Such a system can provide a 360-degree field of view at low resolution, along with a more limited field of view, but at a much higher resolution. These two situations roughly correspond to what happens in the human eye, where the so-called peripheral vision (characterized by a wide angle of view and low resolution) coexists with the so-called foveal vision (characterized by high resolution in the vicinity of the focus of attention). The contemporary presence of the two sensors enables stereo image processing. Such sensors could solve autonomous navigation problems in robotics and for surveillance tasks.

Funded by a grant from the EU within the "Marie Sklodowska Curie" actions (Early Stage Researcher Network), during a four-year research education project ("MIBISOC") in Medical Imaging using Bio-Inspired and soft computing, he explored the use of metaheuristics to optimize contour-based segmentation based on elastic contour models and the use of GPUs to accelerate their implementation.

He subsequently applied evolutionary and swarm intelligence-based methods to accelerate the detection of relevant variable sets in complex systems, or more generally, to model and analyze phenomena described by complex systems with many interacting variables.

Besides receiving funding from the EU, he carried out his research within projects financed by the Italian Ministry of Education, University, and Research (MIUR), the National Research Council (CNR), the Italian Space Agency (ASI), and the National Energy Agency (ENEA). Finally, he managed research contracts with firms in the province of Parma or Emilia Romagna, either through direct agreements or within regional technology transfer programs. Among these, he has co-managed a project funded by the Italian Railway Network Society (RFI) to develop an inspection system for train pantographs based on computer vision. This project generated a patent, whose rights were acquired by a UK-based multinational corporation (CAMLIN Group), whose Italian branch (CAMLIN Italy) has further developed the early prototype to an industrial product level.

Since 1985, he has been a member of the IEEE, of which he is currently a Senior Member. He is a member of the IEEE Computational Intelligence Society's Task Force on Evolutionary Computer Vision and Image Processing, as well as the ACM SIGEVO (Special Interest Group on Evolutionary Computation).

He has been a member of the Managing Board and secretary (2006-2007) of the Italian Association for Artificial Intelligence (AI*IA), a member of the Advisory Board of Perada, the UE Coordination Action on Pervasive Adaptation, and a member of the SPECIES (Society for the Promotion of Evolutionary Computation in Europe and Surroundings) advisory board. Within SPECIES, he currently serves as a member of the advisory board for the EvoApplications conference.

From 1999 to 2018, he chaired a yearly European event dedicated to evolutionary computation applications in image analysis and signal processing, initially as a workshop and later in its current form as a track of the EvoApplications conference, part of the EvoStar multiconference. In 2001 and 2002, he served as General Chair of EvoWorkshops, now known as EvoApplications, the European joint conference on the Applications of Evolutionary Computation.

He was the chairman of the workshop on Evolutionary Computation, held at ECAI, the European Conference on Artificial Intelligence, in 2006, and the founder and chair of GSICE, the Italian Workshop on Evolutionary Computation, in 2005, and chairman in 2006, before the event became the WIVACE (earlier Italian, now International) Workshop on Evolutionary Computation, Artificial Life, and Complex Systems. He was among the organizers of the EvoNet Summer School on Evolutionary Computation in 2003 in Parma, Italy, and SECEVITA, the Summer School on Evolutionary Computation and Artificial Life, in 2007 in Samperi, Italy. In Parma, he also chaired the organizing Committee of WIVACE 2012, Evostar 2018, and WIVACE 2018.

From 2005 to 2020, he co-chaired MedGEC, a workshop on the medical applications of evolutionary computation, and from 2022 to 2025, he co-chaired ECXAI, the workshop on Evolutionary Computation for Explainable AI. Both events were held concurrently with the Genetic and Evolutionary Computation Conference (GECCO).

He co-edited the special issues of the EURASIP Journal of Applied Signal Processing (2003), Pattern Recognition Letters (2006), Evolutionary Computation (2008), and Applied Soft Computing (2020) dedicated to "Genetic and Evolutionary Computation for Image Analysis, Signal Processing, and Pattern Recognition". He also co-edited a special issue of Frontiers in Artificial Intelligence on "Ethical design of artificial intelligence-based systems for decision making" (2023), and a two-part special issue of the ACM Transactions on Evolutionary Learning and Optimization on Explainable AI in Evolutionary Computation (2024-2025)

He has been the Editor-in-Chief of the Journal of Artificial Evolution and Applications from 2007 to 2009.

He serves as a reviewer for many international journals (in particular, he reviewed more than 110 papers for the IEEE Transactions on Evolutionary Computation, and as a member of the scientific committees of several international conferences, among which: Evostar, Genetic and Evolutionary Computation Conference (GECCO), IEEE Conference on Evolutionary Computation (CEC), International Joint Conference on Artificial Intelligence (IJCAI). 

In 2009, he received the "Evostar award" from SPECIES in recognition of his outstanding contributions to Evolutionary Computation.

His scientific output comprises over 180 papers published in international journals and conferences.

As of August 2025, his H-Index is 34 (Google) / 26 (Scopus).