首页|Automated Tongue Segmentation Based on 2D Gabor Filters and Fast Marching
Automated Tongue Segmentation Based on 2D Gabor Filters and Fast Marching
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
In this paper， we propose a novel automated tongue segmentation scheme which can well address the edge enhancement and the initialization problem of the tongue body contour。 First， according to the grey level distribution of the contour， we propose a 2D Gabor magnitude - based detector for the enhancement of the contour of tongue body。 Second， to cope with the discontinuity of edge detection results， we select two stable segments of the tongue body contours and use the fast marching method to obtain the closed tongue body contour。 Moreover， gradient vector flow (GVF) snake is used to obtain the final segmentation result and an augmented Lagrangian method is adopted for fast computation of GVF field。 Qualitative and quantitative comparisons further verify the superiority of the proposed method for the segmentation of tongue body。
NETL
