Change and anomaly detection problems are ubiquitous in science and engineering. The prompt detection of changes or anomalous patterns is often a primary concern, as they provide precious information for understanding the dynamics of a monitored process and for activating suitable countermeasures. Changes, for instance, might indicate an unforeseen evolution of the process generating the data, or a fault in a machinery. Anomalies are typically the most informative regions in an image (e.g., defects in images used for quality control) or the most relevant patterns in a time series (e.g., arrhythmias in ECG tracing) or data stream (e.g. frauds in credit card transactions). Not surprisingly, detection problems in datastreams / time series / images have been widely investigated in the image analysis and pattern recognition communities and are key in application scenarios ranging from quality inspection to health monitoring.
The tutorial presents a rigorous formulation of the change and anomaly-detection problems that fits many signal/image analysis techniques and applications. The tutorial describes in detail the most important approaches in the literature, following the machine-learning perspective of supervised, semi-supervised and unsupervised monitoring tasks. Special emphasis will be given to detection methods based on learned models, which are often adopted to handle images and signals. In particular, these will be divided into traditional models (including autoencoders, learned projections and dictionaries yielding sparse representations) and deep learning models (including CNNs, deep-one-class classifiers and deep generative models)
The tutorial is accompanied by various examples where change/anomaly detection algorithms are applied to solve real world problems. These include health applications (ECG monitoring in wearable devices), quality control (image analysis solutions to detect defects and anomalous patterns in industrial manufacturing), and fraud detection in a datastream of credit card transactions.
Giacomo Boracchi is an Associate Professor of Computer Engineering at Dipartimento di Elettronica, Informazione e Bioingegneria of the Politecnico di Milano, where he also received the Ph.D. in information technology (2008), after graduating in Mathematics (Università Statale di Milano, 2004). His research interests concern image processing and machine learning, and in particular image restoration and analysis, change/anomaly detection, domain adaptation. Since 2016 he has been teaching PhD courses concerning image processing and classification in Politecnico di Milano and Tampere University of Technology (Finland).
Since 2015 he is leading industrial research projects concerning algorithms for X-ray inspection systems for airport security (the algorithm developed with Gilardoni Raggi X has passed European standard C1 for baggage inspection), automatic quality inspection systems for monitoring silicon wafer production (the system developed with STMicroelectronics is currently analyzing wafer production over different sites), and outlier detection in web-sessions (in collaboration with Cleafy).
He has published more than 70 papers in international conferences and journals and he is currently associate editor for IEEE Transactions on Image Processing and IEEE Computational Intelligence Magazine. In 2015 he has received an IBM Faculty Award, in 2016 the IEEE Transactions on Neural Networks and Learning Systems Outstanding Paper Award and in 2017 the Nokia Visiting Professor Scholarship. He has held tutorials in major IEEE conferences: ICIP 2020, ICASSP 2018 and IJCNN 2017, 2018.
Diego Carrera, PhD, System Research Applications, STMicroelectronics.
Diego Carrera graduated Mathematics at Università degli Studi di Milano in 2013 and received the Ph.D. in Information Technology in 2018. In 2015 he has been visiting researcher at the Tampere University of Technology. Currently he is an Application Development Engineer at STMicroelectronics, where he is developing quality inspection systems to monitor the wafer production. His research interests are mainly focused on unsupervised learning algorithms, in particular change detection in high dimensional datastreams, anomaly detection in signal and images and domain adaptation. He held a tutorial on anomaly detection in images in ICIAP 2019.