Abstract | ||
---|---|---|
This work presents a method for the segmentation of optical coherence tomography images of the retina. Before segmenting the tomography, anisotropic diffusion is applied to reduce noise, but preserve the relevant edges. Afterward, the intensity profile of the images is analyzed to extract an initial approximation for the segmentation of three bands within the retina. Finally, a combination of attraction and regularization terms is used to refine the segmentation by fitting the limits of the bands to the highest gradients and smoothing their shapes to make them more regular. From the bands extracted in the different slices of the tomography, a three-dimensional reconstruction is performed for a better visualization of the results. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1007/978-3-319-74727-9_34 | COMPUTER AIDED SYSTEMS THEORY - EUROCAST 2017, PT II |
Field | DocType | Volume |
Anisotropic diffusion,Computer vision,Optical coherence tomography,Visualization,Segmentation,Computer science,Filter (signal processing),Tomography,Smoothing,Regularization (mathematics),Artificial intelligence,Machine learning | Conference | 10672 |
ISSN | Citations | PageRank |
0302-9743 | 0 | 0.34 |
References | Authors | |
2 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Miguel Alemán-Flores | 1 | 69 | 12.06 |
Rafael Alemán-Flores | 2 | 0 | 0.34 |