Adaptive Medical Image Segmentation Algorithm Combined with DRLSE Model

Abstract

The distance regularized level set evolution model proposed recently which uses Gaussian filter to reduce the image noise will yield blurred image edges. Besides, the model still can’t segment adaptively. To slove these problems, we adopt regularized P-M equation filter which can remove the noise while preserve edge information. On the other hand, gradient informarion which depends on inward and outward area's curve is used to improve the direction of the unit normal vector. As a result, the curve can realize adaptive evolution inwards or outwards. In the end, the modified algorithm is applied to accurately extract the medical image's contour, and greatly reduce computational cost.