【摘要】 将Chan和Vese提出的基于Mumford-Shah模型的图像分割方法(C-V方法)由灰度分割扩展到彩色分割,并给予一定改进。提出了一种可以有效分割彩色图像的彩色C-V方法,使得图像中的色度信息得到充分利用;对原始C-V模型进行改进,增强了全局特性,同时考虑到人眼视觉感知,采用合适的色彩空间;将该方法应用于印刷网点图像分割。实验表明,彩色C-V方法用于分割边缘模糊和噪声污染较为严重的网点图像时,效果大大优于传统的分割方法。更多还原

【Abstract】 A new color C-V method is proposed,which makes an extension of the scalar C-V method to the vector-valued case while giving some amelioration to C-V method.This paper proposes a color C-V method which makes sufficient use of color information in images so that it segments color images efficiently.It makes some amelioration to the original C-V method in order to enhance the globality,and considering the characteristics of human vision,chooses the appropriate color space.This paper applies the method to the segmentation of halftone images.The result of experiments show that color C-V method is far better than traditional methods when it transacts halftone images within fuzzy edges and severe pollution of noise.更多还原