Automatic extraction and tracking of the tongue contours

Publisher Website
Google Scholar
Abstract
Computerized analysis of the tongue surface movement can provide valuable information to speech and swallowing research. Ultrasound technology is currently the most attractive modality for the tongue imaging mainly be- cause of its high video frame rate. However, problems with ultrasound imaging such as noise and echo artifacts, refrac- tions, and unrelated reflections pose signicant challenges for computer analysis of the tongue images and hence spe- cic methods have to be developed. This paper presents a system that is developed for auto- matic extraction and tracking of the tongue surface move- ments from ultrasound image sequences. The ultrasound images are supplied by the Head and Transducer Support System (HATS), which was developed in order to x the head and support the transducer under the chin in a known position without disturbing speech. In this work, we pro- pose a novel scheme for the analysis of the tongue images using deformable contours. We incorporate novel mecha- nisms to i) impose speech related constraints on the defor- mations, ii) perform spatiotemporal smoothing using a con- tour postprocessing stage, iii) utilize optical flow techniques to speedup the search process, and iv) propagate user sup- plied information to the analysis of all image frames. We tested the system's performance qualitatively and quantitatively in consultation with speech scientists. Our system produced contours that are within the range of man- ual measurement variations. The results of our system are extremely encouraging and the system can be used in prac- tical speech and swallowing research in the eld of Otolaryn- gology.

This publication has 24 references indexed in Scilit: