Fast edge integration based active contours for color images
Computers and Electrical Engineering
Geometric active contours without re-initialization for image segmentation
Pattern Recognition
A geometric active contour model without re-initialization for color images
Image and Vision Computing
Segmentation of abdominal organs incorporating prior knowledge in small animal CT
ISVC'10 Proceedings of the 6th international conference on Advances in visual computing - Volume Part III
| Hi-index | 0.00 |
Geodesic active contour methods are powerful numerical techniques for image segmentation and analysis. However the edge detector in the model ensures that the data on both sides of the contour is as dissimilar as possible, it cannot deal with the requirement that also the interior of a region should be as homogeneous as possible. In this paper, we present a new region based geodesic active contour method. The new method gives a global view of the boundary information within the image. The method here proposed is particularly well adapted to situations where edges are weak and overlap, and the curve initialization is a very bad guess. A number of experiments on CT medical images were performed to evaluate the new method. The experimental results demonstrate the reliability and efficiency of this new method.