Automatic extraction of blood vessels in the retinal vascular tree using multiscale medialness
| Autor: | Mariem Ben Abdallah, Jihene Malek, Ahmad Taher Azar, Karl Krissian, Philippe Montesinos, Hafedh Belmabrouk, Julio Esclarín Monreal |
|---|---|
| Přispěvatelé: | Université de Monastir - University of Monastir (UM), Benha University (BU), Laboratoire de Génie Informatique et Ingénierie de Production (LGI2P), IMT - MINES ALES (IMT - MINES ALES), Institut Mines-Télécom [Paris] (IMT)-Institut Mines-Télécom [Paris] (IMT), Université de Las Palmas de Gran Canaria [Espagne] (ULPGC) |
| Rok vydání: | 2014 |
| Předmět: |
Hessian matrix
lcsh:Medical physics. Medical radiology. Nuclear medicine lcsh:Medical technology Article Subject Computer science Image map lcsh:R895-920 contours computer.software_genre Image (mathematics) symbols.namesake Radiology Nuclear Medicine and imaging Point (geometry) Eigenvalues and eigenvectors Distance transform business.industry Pattern recognition similarity measures Tree (data structure) lcsh:R855-855.5 [INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] symbols Artificial intelligence Data mining Scale (map) business computer Research Article |
| Zdroj: | International Journal of Biomedical Imaging International Journal of Biomedical Imaging, Vol 2015 (2015) International Journal of Biomedical Imaging, Hindawi Publishing Corporation, 2015, 2015, pp.1-16. ⟨10.1155/2015/519024⟩ |
| ISSN: | 1687-4188 1687-4196 |
| DOI: | 10.1155/2015/519024⟩ |
| Popis: | International audience; We propose an algorithm for vessel extraction in retinal images. The first step consists of applying anisotropic diffusion filtering in the initial vessel network in order to restore disconnected vessel lines and eliminate noisy lines. In the second step, a multiscale line-tracking procedure allows detecting all vessels having similar dimensions at a chosen scale. Computing the individual image maps requires different steps. First, a number of points are preselected using the eigenvalues of the Hessian matrix. These points are expected to be near to a vessel axis. Then, for each preselected point, the response map is computed from gradient information of the image at the current scale. Finally, the multiscale image map is derived after combining the individual image maps at different scales (sizes). Two publicly available datasets have been used to test the performance of the suggested method. The main dataset is the STARE project's dataset and the second one is the DRIVE dataset. The experimental results, applied on the STARE dataset, show a maximum accuracy average of around 94.02%. Also, when performed on the DRIVE database, the maximum accuracy average reaches 91.55%. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Plný text