Vocal folds paralysis detection using an adapted block matching algorithm

Abstract

This paper presents the study of vocal videostroboscopic recordings to detect vocal folds paralysis using a combination of segmentation and block matching techniques. This approach involves three process steps: 1) a pre-processing stage 2) the segmentation of the glottal area is made analyzing the image textures applying Gabor filtering and 3) an adapted version of exhausted block matching algorithm is used. The adaptation consists on for each frame a personalized block size and search window is applied due to segmented region features. Finally, the results show that our proposal works correctly to detect automatically movement abnormalities of the vocal folds.