A Computational Approach to Edge Detection
IEEE Transactions on Pattern Analysis and Machine Intelligence
Fronts propagating with curvature-dependent speed: algorithms based on Hamilton-Jacobi formulations
Journal of Computational Physics
Machine Vision and Applications
A viscosity solutions approach to shape-from-shading
SIAM Journal on Numerical Analysis
A fast level set method for propagating interfaces
Journal of Computational Physics
Global Minimum for Active Contour Models: A Minimal Path Approach
International Journal of Computer Vision
Numerical schemes for the Hamilton-Jacobi and level set equations on triangulated domains
Journal of Computational Physics
Optimal Algorithm for Shape from Shading and Path Planning
Journal of Mathematical Imaging and Vision
An $\cal O(N)$ Level Set Method for Eikonal Equations
SIAM Journal on Scientific Computing
Fast Extraction of Tubular and Tree 3D Surfaces with Front Propagation Methods
ICPR '02 Proceedings of the 16 th International Conference on Pattern Recognition (ICPR'02) Volume 1 - Volume 1
Short note: O(N) implementation of the fast marching algorithm
Journal of Computational Physics
A modified fast marching method
SCIA'03 Proceedings of the 13th Scandinavian conference on Image analysis
Image segmentation of cervical vertebra in X-ray radiographs using the curve fitting strategy
Proceedings of the 2011 ACM Symposium on Applied Computing
IEEE Transactions on Pattern Analysis and Machine Intelligence
Hi-index | 0.00 |
We develop an effective method for improving the segmentation result based on the Multi-Stencils Fast Marching method (MSFM). In MSFM, the gradient information of the image plays a vital role for calculating edges. It is straightforward to obtain the edge of good quality images; however, MSFM may not have robust edge maps available for images with spurious edges. Thus, a special multi-direction circumscribed circle filter is proposed to calculate the image gradient information which is then used in the MSFM. Using the new gradient information, better image contours can be obtained with MSFM. The size of the radius used in our circle filter is constant even the standard deviation of zero-mean Gaussian noise changes while the parameters of mean filter and Canny filter for gradient computation have to be correctly selected according to different noisy images. Our proposed method shows that it is effective through the experiments of image segmentation.