User-steered image segmentation paradigms: live wire and live lane
Graphical Models and Image Processing
European Portuguese MRI based speech production studies
Speech Communication
PROPOR'12 Proceedings of the 10th international conference on Computational Processing of the Portuguese Language
Hi-index | 0.00 |
The study of dynamic aspects of speech production in Portuguese is very important to characterize vowel nasalization. In this context, the analysis of velum movement remains a challenging task and only a few studies present articulatory descriptions of Portuguese nasal vowels. Advances in real-time MRI (magnetic resonance imaging) allow the acquisition of vocal tract images with reasonable spatial and temporal resolution to enable observation and quantification of articulatory movements. The resulting data consists of large image sequences and the structures of interest (e.g., oral cavity) have to be identified (segmented) throughout to enable analysis which can be a time consuming task. This article presents a segmentation tool for real-time MR image sequences of the oral and nasal cavities. The proposed tool has been implemented using MevisLab and provides features for the analysis of the resulting data.