BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Simultaneous Image Segmentation and Registration and Applications 
 - Weihong Guo (Case Western Reserve University)
DTSTART:20171212T143000Z
DTEND:20171212T153000Z
UID:TALK96490@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:Image segmentation and registration play active roles in machi
 ne vision and  image analysis.  In particular\, image registration helps s
 egmenting images when they have low contrast and/or partial missing inform
 ation.  We explore the joint problem of segmenting and registering a templ
 ate (e.g. current) image given a reference (e.g. past) image. We solve the
  joint problem by minimizing a functional that integrates Geodesic Active 
 Contours and Nonlinear Elastic registration. The  template image is modele
 d as a hyper-elastic material (St. Venant-Kirchho&#11\; model) which under
 goes  deformations under applied forces. To segment the deforming template
 \, a two-phase level set based  energy is introduced together with a weigh
 ted total variation term that depends on gradient features of  the deformi
 ng template. This particular choice allows for fast solution using the dua
 l formulation of the  total variation. This allows the segmenting front to
  accurately track spontaneous changes in the shape of  objects embedded in
  the template image as it deforms. To solve the underlying registration pr
 oblem we  use gradient descent and adopt an implicit-explicit method and u
 se the Fast Fourier Transform.  This is a joint work with former PhD stude
 nt Thomas Atta-Fosu.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR