Glottal space segmentation from motion estimation and gabor filtering

Abstract

Obtaining the glottal space segmentation is essential to characterize morphological disorders of vocal folds. In this study, the tested images are been acquired by direct optical inspection of the glottis using an endoscope and most of them are very poor quality. The application of motion estimation is very useful to segment the vocal folds endoscopic videos without user interaction. This approach involves three process steps: 1) Wiener motion estimator--to shift the measurement the next frame regarding to the current frame, and look for similarities between them. The best matching will accurate a shift equal to the displacement vector of the object; 2) Segmentation using motion estimation results and applying Gabor filtering; 3) Experimental results to demonstrate that the proposed method is effective. Our proposal works correctly with 95% of database test videos and it shows a great advance in design, and in the nearby future, a complete method to diagnose vocal folds pathologies.

Laryngeal Image Processing of Vocal Folds Motion

“…In [97], the authors combine an anisotropic diffusion with an FFT-based band pass filter in order to obtain a smoother image without losing edge information (second row of Figure 7). In [83] a Lagrange interpolation is combined with a Gaussian filter in order to smooth the images, reduce noise and eliminate unwanted details.…”

“…A similar strategy has been implemented in [72], which apply an algorithm based on differences between consecutive frames. Other works [11,97,98] use motion estimation techniques to compute the ROI based on the fact that the region with the most salient motion features corresponds to the vocal folds.…”

“…There are other approaches that combine different image-processing techniques [75,86,89,91,94,97], use feature extraction and training [95], or make use of more recent paradigms such as Deep Neural Networks [96]. For instance, [89] present an automatic glottis segmentation approach using deformable shape models and region growing.…”

Computer based quantification of normal and pathological vocal folds phonatory processes from laryngovideostroboscopy

“…There have been several automatic segmentation methods of the vocal folds images proposed [4]- [6]. Many of the developed methods and algorithms are designed for specific image recording conditions and work  properly only for local databases containing videos collected in a particular institute, hospital or health center.…”

Computer aided tool for diagnosis of ENT pathologies using digital signal processing of speech and stroboscopic images

“…To include different digital signal processing algorithms developed by the authors in previous works in the tool (Méndez et al Mendez et al [2009]; Mendez et al [2012]), (García et al [2009]).…”