基于水平集的曲线演化在图像分割中的应用

Application of Curve Evolution Based on Level Set in Image Segmentation

图像分割是图像处理中的一项基础工作,一般有基于边界和基于区域的分割方法。随着水平集理论的提出,主动轮廓模型尤其是C-V模型,在图像分割中的应用得到迅速发展。多相C-V模型,融合形状先验、纹理等信息的模型也被提出并得到较好的应用。将C-V模型应用于图像分割中,对合成图像和真实图像的分割实验,以及与其他分割方法的对比实验证明了该模型在图像分割中的有效性。

Image segmentation is one of the most fundamental tasks in image processing.Two segmentation methods which are edge based and region based are used mostly in image segmentation.With the presentation of Level Set,active contour models,especially the C-V model,have developed rapidly in the image segmentation application.Besides,multiphase C-V model and models incorporated by shape prior,texture and other prior information have also been proposed.The C-V model was applied to image segmentation in this paper.Its effectiveness was testified both through segmentation experiments on synthetic and real images and comparison experiments to other segmentation methods.