Geodesic Tracking of Retinal Vascular Trees with Optical and TV-Flow Enhancement in SE(2)

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review


Retinal images are often used to examine the vascular system in a non-invasive way. Studying the behavior of the vasculature on the retina allows for noninvasive diagnosis of several diseases as these vessels and their behavior are representative of the behavior of vessels throughout the human body. For early diagnosis and analysis of diseases, it is important to compare and analyze the complex vasculature in retinal images automatically. In previous work, PDE-based geometric tracking and PDE-based enhancements in the homogeneous space of positions and orientations have been studied and turned out to be useful when dealing with complex structures (crossing of blood vessels in particular). In this article, we propose a single new, more effective, Finsler function that integrates the strength of these two PDE-based approaches and additionally accounts for a number of optical effects (dehazing and illumination in particular). The results greatly improve both the previous left-invariant models and a recent data-driven model, when applied to real clinical and highly challenging images. Moreover, we show clear advantages of each module in our new single Finsler geometrical method.

Originele taal-2Engels
TitelScale Space and Variational Methods in Computer Vision - 9th International Conference, SSVM 2023, Proceedings
RedacteurenLuca Calatroni, Marco Donatelli, Serena Morigi, Marco Prato, Matteo Santacesaria
Plaats van productieCham
UitgeverijSpringer Nature
Aantal pagina's13
ISBN van elektronische versie978-3-031-31975-4
ISBN van geprinte versie978-3-031-31974-7
StatusGepubliceerd - 2023

Publicatie series

NaamLecture Notes in Computer Science
UitgeverijSpringer Cham
ISSN van geprinte versie0302-9743
ISSN van elektronische versie1611-3349


Duik in de onderzoeksthema's van 'Geodesic Tracking of Retinal Vascular Trees with Optical and TV-Flow Enhancement in SE(2)'. Samen vormen ze een unieke vingerafdruk.

Citeer dit