Research (21)
Lunedì, 19 Novembre 2018 11:25
Hand-designed descriptors vs. pre-trained convolutional networks: a comparison of two strategies for colour texture classification
Scritto da Bianconi FrancescoELG 10 at 14:00 Tuesday 20th November 2018.
City, University of London.
Abstract: Colour texture classification plays a fundamental role in a wide range of applications, as for instance surface inspection and grading, content-based image retrieval, computer-assisted diagnosis and remote sensing. Traditionally, the problem has been approached by manually designing suitable functions capable of extracting meaningful visual features from the input images: the so-called ‘hand-crafted’ paradigm. In recent years, however, the arrival on the scene of Convolutional Neural Networks (`Deep Learning') has brought about dramatic advances in many areas of computer vision, and has the potential to significantly change the approach to texture analysis as well. This seminar will comparatively evaluate – both at a theoretical and experimental level – the hand-crafted and Deep Learning paradigms for colour texture classification. There will be specific focus on the use of pre-trained networks as generic feature extractors for colour textures.
Abstract: Colour texture classification plays a fundamental role in a wide range of applications, as for instance surface inspection and grading, content-based image retrieval, computer-assisted diagnosis and remote sensing. Traditionally, the problem has been approached by manually designing suitable functions capable of extracting meaningful visual features from the input images: the so-called ‘hand crafted’ paradigm. In recent years, however, the arrival on the scene of Convolutional Neural Networks (`Deep Learning') has brought about dramatic advances in many areas of computer vision, and has the potential to significantly change the approach to texture analysis as well. This seminar will comparatively evaluate – both at a theoretical and experimental level – the hand-crafted and Deep Learning paradigms for colour texture classification. There will be specific focus on the use of pre-trained networks as generic feature extractors for colour textures.
Pubblicato in
Research
Lunedì, 02 Luglio 2018 15:15
CT textural and shape features as prognostic indicators for NSCLC
Scritto da Bianconi FrancescoIntense research is being conducted throughout the world to identify biomarkers seen on computed tomography (CT) imaging that can help predict prognoses in patients with non-small-cell lung cancer (NSCLC). Such biomarkers could help improve and personalize treatment plans for patients. A team of radiologists, engineers, and biomedical specialists at the University of Perugia evaluated 30 three-dimensional shape and textural CT-derived features as potential biomarkers predictive of overall survival of 203 NSCLC patients. Find out more.
Pubblicato in
Research
Etichettato sotto
SOFTWIND - PRIN 2015
www.softwind.it
E' on line il sito web del PRIN 2015 finanziato dal MIUR, Progetto di Ricerca di Interesse Nazionale dal titolo "SOFTWIND turbines" Il progetto coordinato a livello nazionale dall'Università di Camerino vede Perugia tra le quattro unità componenti il team di ricerca.
Il gruppo del nostro Ateneo è coordinato dal Prof. Filippo Cianetti e composto inoltre dal Prof. Francesco Castellani. Si occuperanno di modellazione multicorpo, di modellazione orientata al controllo dei generatori e di valutazione del comportamento a fatica di queste complesse macchine.
La finalità del progetto è quella appunto di minimizzare la danneggiabilità dei generatori e delle strutture nel loro complesso controllando l'assetto delle pale durante il loro funzionamento e sviluppando materiali innovativi per la loro realizzazione e sensori altrettanto smart da poterne misurare le prestazioni strutturali durante il loro funzionamento.
Pubblicato in
Research
Domenica, 21 Maggio 2017 11:31
Prospettive occupazionali al top per laureati in Ingegneria
Scritto da Bianconi Francesco
Secondo il nuovo rapporto di Almalaurea il 93,6% dei laureati magistrali in Ingegneria trova occupazione dopo un anno dalla fine degli studi e con lo stipendio più alto di tutti gli altri laureati (€ 1.717 in media netti al mese). Anche i laureati triennali hanno ottimi risultati con il 76% occupati ed uno stipendio medio di € 1.283.
Pubblicato in
Research
Domenica, 05 Marzo 2017 19:09
Computational Colour Imaging Workshop - 2017
Scritto da Bianconi FrancescoWe will be present with one paper and one tutorial at the Computational Colour Imaging Workshop (CCIW'17), to be held in in Milan, Italy, from March 29th through March 31st, 2017. This is an international forum for research in colour image processing, colour image quality assessment, colour vision modelling, and colour image reproduction. The workshop aims at presenting research that advances the state-of-the-art in these and related areas.
Pubblicato in
Research
Martedì, 28 Giugno 2016 10:31
POKer: a partial-order derivative kernel for sequences with alternative symbols
Scritto da Bianconi FrancescoPOKer: a partial-order kernel for sequences with alternative symbols
Dr. Maryam Abdollayhan (Queen Mary, University of London)
Department Of Engineering, aula adiacente Aula Magna, 05 July 2016, 5pm
Pubblicato in
Research
Lunedì, 25 Aprile 2016 16:06
Ranklets: Orientation selective non-parametric features for image analysis
Scritto da Bianconi Francesco
Ranklets: Orientation selective non-parametric features for image analysis
Dr. Fabrizio Smeraldi (Queen Mary, University of London)
Department Of Engineering, Aula Magna, 04 May 2016, 4pm
Dr. Fabrizio Smeraldi (Queen Mary, University of London)
Department Of Engineering, Aula Magna, 04 May 2016, 4pm
Pubblicato in
Research
Etichettato sotto
We're happy to announce that Dr. Fabrizio Smeraldi, Lecturer at Queen Mary, University of London, will be visiting the Department of Engineering for four months starting from March 14th, 2016. Welcome aboard Fabrizio!
Pubblicato in
Research
Etichettato sotto
Texture analysis plays a pivotal role in many machine vision applications including, among the others, product inspection, surface grading, defect detection, content-based image retrieval, computer-assisted diagnosis and remote sensing. This seminar will introduce some basic concepts about texture analysis and will focus on a class of methods known as 'Histogram of Equivalent Patterns'. The concepts treated are accessible to a general technical audience and only require some basic knowledge of image processing. Date: Sep. 22, 2015 - 15:00hrs Queen Mary, University of London Department of Electical and Electronic Engineering Room: ITL Top Floor Seminar Room Speaker: Francesco Bianconi Università degli Studi di Perugia Department of Engineering
Pubblicato in
Research
The Workshop on Color and Texture in Material Recognition will take place in Genoa, Italy, in conjunction with the 18th International Conference on Image Analysis and Processing (ICIAP), Sept. 7-11, 2015. The workshop will cover different areas of image processing including colour and texture analysis, material perception, remote sensing, medical imaging and industrial inspection. Deadline for paper submission: June 3, 2015.
