سال انتشار: ۱۳۹۰

محل انتشار: هفتمین کنفرانس ماشین بینایی و پردازش تصویر

تعداد صفحات: ۵

نویسنده(ها):

Morteza Jalalat Vakilkandi – ECE Dept., Isfahan University of Technology-MISP Center, Medical university of Isfahan 2Student Member, IEEE
Mohammad Reza Ahmadzadeh – ECE Dept., Isfahan University of Technology-MISP Center, Medical university of Isfahan
Rasoul Amirfattahi – ECE Dept., Isfahan University of Technology-MISP Center, Medical university of Isfahan
Saeid Sadri – ECE Dept., Isfahan University of Technology-MISP Center, Medical university of Isfahan

چکیده:

Correct discern of blood vessels performs a significantrole in a number of clinical applications, especially inangiographic image analysis. In this paper, we propose a newsimple and fast fabrication of DDFB (called FDFB) withoutnecessity of downsampling and resampling employed in DDFB,which is time consuming, especially if the directions are sizeable.Also, in this paper we discuss on parameters of vessel filter inmore detail and propose a proper range for parameters selectionto achieve accurate vessel enhancement. One major problem inautomatic segmentation is the presence of background non-vesselstructures; we use a morphological-based algorithm to removethese, after vessel enhancement. Another problem with DDFB isthe ringing artefact; smoothed FDFB unravels the later. Theexperimental results show that the proposed FDFB and vesselenhancement framework work well on angiograms