"Fast registration by boundary sampling and linear programming" charla de Jan Kybic de la Czech Technical University el día 26 de septiembre a las 17:00 University Institute for Computing Research

Ir a contenido Ir a Estudios, Gobernanza y organización
Logo UA
Realizar búsqueda
Menú
Research Institutes
Logo University Institute for Computing Research   University Institute for Computing Research
University Institute for Computing Research
IUII
NEWS

"Fast registration by boundary sampling and linear programming" charla de Jan Kybic de la Czech Technical University el día 26 de septiembre a las 17:00

23/09/2019


Fast registration by boundary sampling and linear programming
Jan Kybic de la Czech Technical University.

  • Fecha: 26 de septiembre
  • Hora: 17:00h 
  • Lugar: sala de posgrado del departamento de Ciencia de la Computación e I.A. (Edificio Politécnicas 3, primera planta, tercer pasillo).

Resumen:

Image registration is one of the key image analysis tasks, especially in biomedical imaging. However, accurate image registration methods are often slow and this problem is exacerbated by the steadily increasing resolution of today's acquisition methods. In my talk, I will present our approach, how image registration can be accelerated.

First, we take advantage of the fact that image registration is mostly driven by image edges. We take this idea to the extreme. We approximate the similarity criterion by sampling only a small number of sparse keypoints and consider only  normal displacements. Furthermore, we simplify images by segmenting them first.  The segmentation can be performed jointly and alternated with the registration steps. 

We  create a piecewise linear convex approximation of the individual contributions. We obtain a linear program for which a global optimum can be found very quickly by standard algorithms. 
The linear program formulation also allows for an easy addition of regularization and trust-region bounds. We have tested the approach for affine and B-spline transformation representation but any linear model can be used. Larger deformations can be handled by multiresolution. We show that our method is much faster than pixel-based registration, with only a small loss of accuracy. In comparison to standard keypoint based registration, our method is applicable even if individual keypoints cannot be reliably identified and matched, and it is at least one magnitude faster than standard pixel-based approaches.



  

News