Guang Li, Shouhua Luo, Tao Jiang, Ning Gu, and Yuling Yan
High-speed digital imaging, Vocal fold vibration, Glottis, Automatic contour tracking, Region of interest, Threshold segmentation
High-speed digital imaging (HSDI) is an emerging modality for endoscopic laryngeal imaging. HSDI has a high acquisition rate sufficient to resolve the fast phonatory vocal fold vibrations. The characteristics of the vocal fold vibration play an important role in the clinical assessment of voice disorders as well as understanding the mechanism of phonation. It is therefore of great interest for voice clinicians to measure and quantify information on the vocal fold motion during phonation. In this paper, a semi-automatic algorithm is introduced for efficient and accurate glottis segmentation. This method consists of two major steps: first, an automatic tracking of the glottis contour and second, a global threshold segmentation. The reliability and accuracy of the algorithm are evaluated using representative clinical HSDI recordings.
Important Links:
Go Back
