The Signal Processing Laboratory at ICS-FORTH, aspires to be at the forefront of signal processing with fundamental work on image, audio, and speech signal processing and learning theory. Our research activities rely on the pillars of non-Gaussian statistics, sinusoidal modeling, sparse representations, and compressed sensing, and our aim is to apply state of the art techniques to a wide range of real world problems. Representative applications include imaging and video coding, distributed signal and data management over sensor networks, immersive audio and multichannel audio coding, speech recognition and enhancement.

Imaging

We synthesize, extend, and apply theoretical breakthroughs on Machine Learning, Compressed Sensing, and Sparse Representations, to emerging Imaging applications. Our key areas of interest range from Active Range Imaging, to Land-cover classification for remote sensing, and Computational Photography.

Read More

Distributed Sensor Networks

We have a distinct focus on signal acquisition/representation, distributed processing, and context retrieval from heterogeneous sensing devices, as well as on autonomic networking algorithms for limited bandwidth, limited power, and limited capacity wireless sensors, which serve the design paradigm of IoT and CPS.

Read More

Audio & Speech

We perform cutting edge research, featuring microphone array signal processing, wireless acoustic sensor networks, multichannel audio coding, immersive audio capturing and reproduction, sound source separation, speech enhancement for improved intelligibility, and robust automatic speech recognition.

Read More

Featured Publications

E. Doutsi, L. Fillatre, M. Antonini, and P. Tsakallides, “Dynamic Image Quantization Using Leaky Integrate-and-Fire Neurons,” IEEE Transactions on Image Processing, vol. 30, pp. 4305-4315, 2021, doi: 10.1109/TIP.2021.30701932021.

E. Troullinou, G. Tsagkatakis, S. Chavlis, G. Turi, W.-K. Li, A. Losonczy, P. Tsakallides, and P. Poirazi, “Artificial Neural Networks in Action for an Automated Cell-Type Classification of Biological Neural Networks,” IEEE Transactions on Emerging Topics in Computational Intelligence, vol. 5, no. 5, pp. 755-767, Oct. 2021, doi: 10.1109/TETCI.2020.3028581.

A. Aidini, G. Tsagkatakis, and P. Tsakallides, “Tensor Decomposition Learning for Compression of Multidimensional Signals,IEEE Journal of Selected Topics in Signal Processing, vol. 15, no. 3, pp. 476-490, April 2021, doi: 10.1109/JSTSP.2021.3054314.

R. Stivaktakis, G. Tsagkatakis, B. Moares, F. Abdalla, J. L. Starck, and P. Tsakallides, “Convolutional Neural Networks for Spectroscopic Redshift Estimation on Euclid Data,” IEEE Transactions on Big Data, vol. 6, no. 3, pp. 460-476, September 2020, doi: 10.1109/TBDATA.2019.2934475.

G. Tsagkatakis, A. Aidini, K. Fotiadou, M. Giannopoulos, A. Pentari, and P. Tsakallides, “Survey of Deep-Learning Approaches for Remote Sensing Observation Enhancement,” MDPI Sensors, 2019, 19, 3929; doi:10.3390/s19183929.

R. Stivaktakis, G. Tsagkatakis, and P. Tsakallides, “Deep Learning for Multi-Label Land Cover Scene Categorization Using Data Augmentation,” IEEE Geoscience and Remote Sensing Letters, vol. 16, no. 7, pp. 1031-1035, July 2019, doi: 10.1109/LGRS.2019.2893306.