سال انتشار: ۱۳۹۰
محل انتشار: هفتمین کنفرانس ماشین بینایی و پردازش تصویر
تعداد صفحات: ۵
F. Mokhayeri – Islamic Azad University, Mashhad BranchMashhad, Iran
S. Toosizadeh – Islamic Azad University, Mashhad BranchMashhad, Iran
M-R. Akbarzadeh-T – Ferdowsi University of MashhadMashhad, Iran
In this paper, a hybrid approach for the pupildiameter (PD) measurement from the eye video is described. Theproposed algorithm is composed of five main stages; dataacquisition, pre-processing, edge detection, edge linking, anddiameter computation. In this approach, noise reduction isperformed using the type-1 fuzzy system, eye region is detectedusing the genetic algorithm (GA), and an effective procedure foredge detection based on a hybrid combination of interval type-2fuzzy logic system (IT2FS) and particle swarm optimization(PSO) is presented that PSO is used for parameters optimizationof IT2FS membership function. In order to evaluate theperformance of proposed algorithm, a pupil videos dataset isconstructed. The Proposed algorithm is tested on this dataset.The experimental results show that the combination of IT2FSand PSO can be considered as a good alternative for the edgedetection, and proposed approach can be considered as a generalmethod of pupil diameter measurement.