School of Electronic Engineering
Signal and Image Processing
Research in Signal and Image Processing is focused on three main research streams: Image Processing and Analysis (including computer vision and medical imaging), Digital Media (including multi-modal digital media analysis and processing) and Speech (including speech acoustics, synthesis, recognition, speaker transformation and signal processing).
Research areas
- Computer Vision and Medical Imaging: Image processing and analysis for industrial, robotic, computer aided detection and diagnostic biomedical applications. Visual biometrics, virtual reality and image visualisation
- Multi-modal digital media analysis and processing for applications in ambient intelligence, personal health and social media. This targets analysis and indexing of audiovisual data and complementary information sources
- Speech synthesis and speech recognition
Image Processing & Analysis Core (Vision Systems Group)
Vision Systems Group (Centre for Image Processing and Image Analysis [CIPA]):
The Vision Systems Group (VSG) is a key university recognised research group. It is the Image Processing and Analysis (IPA) research core in DCU. Our digital and non-digital IPA research programmes relate to issues involved in the acquisition (customer sensor design), processing quantitative analysis, classification, visualisation and systems engineering (integration) for a wide range of computer vision applications. Specifically the group focuses on the issues involved in the automation or semi automation of image feature segmentation, and its associated quantitative analysis, at both a micro and macro level. The core expertise provided by the Vision Systems Group in DCU is in its ability to develop and design novel computer based solutions that will allow the automatic extraction of key image features [specifically from 2D, video, 3D and 4D data sources] with a view to a robust and reliable quantitatively analysis of the key information/data within the scene. The VSL is both a Science Foundation Ireland (SFI) and HEA-PRTLI funded research group. For additional information please refer to the group’s web site at www.vsg.dcu.ie.
Research Interests within Image Processing and Analysis/Computer Vision
Professor P.F. Whelan, B.Eng., M.Eng., Ph.D., C.Eng., MIEE, SMIEEE
- Image processing and quantitative image analysis
- Computer and machine vision
- Medical image analysis
- Computer assisted detection and diagnosis
- 3D Imaging of biological surfaces
Dr. D. Molloy, B.Eng., Ph.D., M.I.E.I., A.M.I.E.E.
- Machine vision (motion analysis)
- Virtual reality applications
- Web-based teaching
- Software engineering
Dr. R. Sadlier, B.Eng., Ph.D, M.I.E.I., M.I.E.E.E.
- Computer and machine vision
- 3D computer graphics
- Medical image analysis
Digital Media
Over the last decade, research activities within the field of digital media have grown out of the School's background in speech coding, audio analysis, computer vision, image and video compression and audiovisual content analysis. In the past, School activities in this area have contributed to the definition of world-wide standards such as ITU-T H.263, ISO MPEG-4 and MPEG-7. Currently, research is focused on mining context and semantic information from digital media by considering capture mechanisms augmented with multiple complementary sensing modalities. Objectives include developing fusion frameworks that handle multiple potentially conflicting data sources and extending the use of pattern recognition techniques to model content semantics using non-traditional sensor input. Applications targeted are in areas as diverse as ambient assisted living, social media and personal health. Given the growing popularity of using sensor network paradigms in such applications, research also focuses on investigating power aware reconfigurable hardware for media processing on next generation media sensor nodes.
Research interests within Digital Media
Dr. N. E. O’Connor, B.Eng., Ph.D., MIEI, MIET, MIEEE
(see also Centre for Digital Video Processing)
- Audiovisual content understanding for extracting context and semantics
- Object, event and activity analysis and modelling
- Content processing and indexing for browsing, searching, alerting, filtering and summarisation
- Power aware reconfigurable hardware for media processing on mobile platforms
Speech Laboratory
The Speech Laboratory (founded in 2004) is an inter-disciplinary team based in the Schools of Electronic Engineering and Computing, bringing together expertise in speech acoustics, synthesis, recognition, speaker transformation and signal processing. Our current work in synthesis covers concatenative and statistical parametric synthesis, with a specific interest in new models of perception of discontinuity and perceptual experiments to validate these models. Our work on speaker transformation ranges over non-linear estimation of speech production models and dimensionality reduction for voice modification. The Lab has two current Ph.D. students and several Masters students.
Research interests in Speech Laboratory
Dr. R. Scaife, B.A., B.A.I., Ph.D., M.ISCA, M.ASA
- Speech acoustics
- Perception of speech discontinuities
- HMM speech synthesis
- Virtual-physical musical instruments
Dr. S. Marlow, B.Sc., Ph.D.
- Audio and video coding and analysis
- Audiovisual indexing and retrieval
- Digital signal processing
Research Laboratories
For laboratory specific homepages, please visit:
