Interactive Contour Detection (Image Segmentation) and Edge Damage Evaluation

Di Yang (ANU)

COMPUTER VISION AND ROBOTICS SERIES HDR Monitoring

DATE: 2011-04-28
TIME: 16:00:00 - 17:00:00
LOCATION: RSISE Seminar Room, ground floor, building 115, cnr. North and Daley Roads, ANU
CONTACT: JavaScript must be enabled to display this email address.

ABSTRACT:
The problem of efficient, interactive contour detection (image segmentation) in still images is of great practical importance in quality control (e.g. edge damage evaluation). Classical contour detection (image segmentation) methods use either texture (colour) information or edge (contrast) information. In this seminar, we will present three existing approaches, namely "Watershed", "GrabCut" and "LevelSet". Through evaluating the three methods, "border matting", "alpha-matte" (transparency) and "reflections" will be discussed, which are three major constraints, limiting performance of boundary detection.
BIO:
Di Yang is a PhD candidate of ANU.