Invited speakers

Patrizio Campisi

Università degli Studi Roma Tre, Roma, Italy

Title: Biometrics and neuroscience: a possible marriage?

Abstract: In everyday life, biometric systems, mainly relying fingerprint, face, and iris, are more and more used as a convenient way to securely recognize people. In addition, in the las few years, an always growing interest towards the use of cognitive biometrics for automatic people recognition is being witnessed.

In this talk we focus on the use of brain signals, and more specifically on electroencephalographic (EEG) signals, as emerging biometric identifiers. Despite some preliminary studies regarding the use of EEG biometrics have already been performed, many fundamental questions still remain open and need a more deep and systematic investigation. Therefore, we take a critical perspective by analyzing the pros and cons related to the use of EEG biometrics, and speculate about the EEG stability in time, the EEG signals discriminative power, the enhanced security EEG systems can provide, the user (in)convenience related to EEG acquisition systems, and the relationship of these characteristics with the employed acquisition protocol (elicitation stimuli, electrodes displacement and number, etc.). With these goals in mind, the state of the art of EEG based biometric systems is critically discussed and the research agenda for deploying EEG biometrics in real life is drafted.

Bart ter Haar Romeny

University of Eindhoven, the Netherlands
Northeastern University, Shenyang, China

Title: Brain-inspired Methods for Retinal Image Analysis

Abstract:Diabetes is reaching epidemic proportions worldwide, especially in Asia due to fast lifestyle changes and genetic factors. A recent study revealed that 11.6% of the Chinese population had diabetes-2 in 2013, the most of any country. Diabetic retinopathy is also the main cause of newly formed blindness in the working population, leading to high societal costs. Early detection is the key to prevention and successful treatment of these forms of blindness. Many cases still go unnoticed and are not treated in time, especially in rural areas.

The RetinaCheck project is a large-scale screening project between different partners in the Netherlands and China to detect early diabetes by changes in high-resolution retinal fundus images by fully automated advanced computer vision detection and analysis techniques. Brain-inspired multi-scale and multi-orientation methods are exploited to enhance, denoise and normalize images, to extract quantitative biomarker features, such as vessel pattern fractal dimension, optic disk rim shape, micro-bleed detection, vessel tortuosity and branching and cross-over detection from multi-orientation scores, multi-orientation vesselness etc. The classification power of the combined features is validated in a clinical setting where many diabetic and ophthalmologic metadata are recorded. Most of the software design is done in Mathematica.

The lecture will give an overview of this international project, and discuss in depth the math and implementation issues of the brain-inspired computer vision methods.

Mario Vento

University of Salerno, Italy

Title: How and Why Graphs are used in Pattern Recognition?

Abstract: The talk presents a historical overview of graph-based methodologies in Pattern Recognition. Graphs have been used extensively in Pattern Recognition in all those cases in which the objects of interest are represented by means of parts suitably connected with each other, so as to obtain somehow a graph.

The use of a graph-based pattern representation induces the need to formulate the main operations required in Pattern Recognition in terms of operations on graphs: classification, usually intended as the comparison between an object and a set of prototypes, and learning, which is the process for obtaining a model of a class starting from a set of known samples, are among the key issues that must be addressed using graph-based techniques.

Forty years have passed since the first papers on this topic appeared in Pattern Recognition literature: a lot of research effort has been devoted to explore this challenging field and some approaches have been meanwhile consolidated. The talk is focused on the most relevant techniques proposed along these years, so as to give the audience an overview of the field.

The overview is aimed at reveal the rationale inspiring the papers published in these years, so as to roughly classify them. Despite the extent of scientific production in this field, it is possible to identify three historical periods, each having its own connotation common to most of the corresponding papers, which are called here as the pure, the impure and extreme periods.