Boundary vector field
The boundary vector field[1][2] (BVF) is an external force for parametric active contours (i.e. Snakes). In the fields of computer vision and image processing, parametric active contours are widely used for segmentation and object extraction. The active contours move progressively towards its target based on the external forces. There are a number of shortcomings in using the traditional external forces, including the capture range problem, the concave object extraction problem, and high computational requirements.
The BVF is generated by an interpolation scheme[1] which reduces the computational requirement significantly, and at the same time, improves the capture range and concave object extraction capability.
The BVF is also tested in moving object tracking and is proven to provide fast detection method for real time video applications.[3]
References
- K.W. Sum and Paul Y.S. Cheung, "Boundary Vector Field for Parametric Active Contours," Pattern Recognition, vol. 40, no. 6, pp. 1635–1645, Jun 2007
- Rafael Verdú-Monedero, Juan Morales-Sánchez, and Luis Weruaga, "Convergence Analysis of Active Contours," Image and Vision Computing, vol. 26, issue 8, pp. 1118–1128, 2008
- N. Lin and B. Hu, "Moving Object Detection and Tracking in Video Sequences Based on Boundary Vector Field," Journal of Computer Applications, vol. 28, Jun 2008
- Pattern Recognition (Journal of the Pattern Recognition Society)
- Image and Vision Computing, Elsevier Journal
- Journal of Computer Applications
 |  |
