BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Lancaster University Faculty of Science and Technology//NONSGML v1.0//EN
BEGIN:VTIMEZONE
TZID:/Europe/London
X-LIC-LOCATION:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;INTERVAL=1;BYDAY=-1SU;BYMONTH=3
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;INTERVAL=1;BYDAY=-1SU;BYMONTH=10
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:93
SUMMARY:Model-based on-line Fault Detection and Image Classification
DESCRIPTION:A seminar by DR. Edwin Lughofer from Johanes Kepler Univeristy, Linz, Austria visiting Fellow of the Royal Society (host: Dr. P. Angelov)\n\nAn On-Line Interactive and Self-Adaptive Image\nClassification Framework\nEdwin Lughofer\nedwin.lughofer@jku.at, http://www.flll.jku.at/people/staff/lughofer.html\nJohannes Kepler University Linz\nAltenbergerstrasse 69, A-4040 Linz, Austria\n\nAbstract\n\nIn many machine vision applications, such as inspection tasks for quality control, an automatic system tries to reproduce\nhuman cognitive abilities. The most efficient and flexible way to achieve this, is to learn the task from a human expert. This\ntraining process involves object recognition methods, adaptive feature extraction algorithms and evolving classifiers. A lot of\nresearch has been done on each of these topics, however, simply plugging all of these methods together does not necessarily lead\nto a working machine vision system.\nIn this talk, a generic self-adaptive image classification framework is presented, focussing on integration issues and on topics\nthat are specific to quality control applications. The framework is shown in Figure 1. The basic components of these framework\nare the following:\n• Generating contrast images by calculating the deviation images to the master\n• Recognizing ROIs (=regions of interest) in the deviation images\n• An adaptive object and aggregated feature extraction component (object features characterize single objects, whereas aggregated\nfeatures characterize whole images)\n• Training of (initial) base classifiers for aggregated and object features (! aggregated and object classifiers) based on off-line\npre-labeled image sets\n• Ensemble Classifiers for resolving contradictory input among different operators\n• Strategy for on-line classification of new images (incorporating object, aggregated and ensemble classifiers)\n• On-line adaptation/evolution of base and ensemble classifiers (based on operator's feedback during on-line mode)\n• Early prediction of success or failure of a classifier\nEach of these components will be addressed during the talk.\n\nAt the end of the talk some results on real-recorded images in various industrial systems will be presented. These results will\ninclude a comparison of object recognition methods as well as off-line and on-line classification accuracies and achievable bounds\n(from base and ensemble classifiers) based on (off-line) pre-labeled data and operator's feedback during on-line operation mode.
DTSTART:20070524T130000
DTEND:20070524T140000
LOCATION:C60 InfoLab21
END:VEVENT
END:VCALENDAR