Scale-Space and Edge Detection Using Anisotropic Diffusion
IEEE Transactions on Pattern Analysis and Machine Intelligence
Mean Shift: A Robust Approach Toward Feature Space Analysis
IEEE Transactions on Pattern Analysis and Machine Intelligence
Diffusion Distance for Histogram Comparison
CVPR '06 Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 1
IEEE Transactions on Knowledge and Data Engineering
Image Segmentation Via Iterative Geodesic Averaging
ICIG '09 Proceedings of the 2009 Fifth International Conference on Image and Graphics
ACM Transactions on Graphics (TOG)
Comparing distributions and shapes using the kernel distance
Proceedings of the twenty-seventh annual symposium on Computational geometry
Image segmentation based on electrical proximity in a resistor-capacitor network
ACIVS'11 Proceedings of the 13th international conference on Advanced concepts for intelligent vision systems
| Hi-index | 0.00 |
In image segmentation, measuring the distances is an important problem. The distance should tell whether two image points belong to a single or, respectively, to two different image segments. Although the Euclidean distance is often used, the disadvantage is that it does not take into account anything what happens between the points whose distance is measured. In this paper, we introduce a new quantity called the energy-transfer proximity that reflects the distances between the points on the image manifold and that can be used in the image-segmentation algorithms. In the paper, we focus especially on its use in the algorithm that is based on k-means clustering. The needed theory as well as some experimental results are presented.