一种新的测地线主动轮廓图像分割方法

测地线主动轮廓模型是一种重要的图像分割方法,它对噪声图的分割效果很大程度上依赖于图像平滑的质量.为了使图像平滑和图像分割有机的结合起来,文中首先利用方向曲率模值提出描述图像平滑度的泛函,推导出一种四阶偏微分方程(PDE)图像降噪模型,它能在有效降噪的同时,保持良好的图像特征.该方法处理结果为分段线性图像,且在目标边缘处梯度存在阶跃.利用降噪结果的这一特点作为图像特征的描述函数,文中提出一种新的测地线主动轮廓(new geodesic active contour)模型.实验表明,新模型轮廓提取能力强、收敛速度快.以文中的降噪模型进行预处理,对基于区域的主动轮廓模型分割效果也有较大的提高.

New method of geodesic active contours for image segmentation

The geodesic active contour method is important for image segmentation,but it cannot segment noisy images effectively.To combine image enhancement and image segmentation as one process,a class of fourth-order partial differential equations (PDEs) is first proposed to optimize the trade-off between noise removal and edge preservation. The time evolution of these PDEs seeks to minimize a cost function, which is an increasing function of the curvature magnitude of the image intensity function. These PDEs attempt to remove noise and preserve edges by approximating an observed image with a piecewise planar image, so the gradient of the processed image is not continuous at the boundary. Then, a novel scheme for the detection of object boundaries is presented. The technique is based on geodesic active contours evolving in time according to intrinsic geometric measures of the laplacian of image, which is infinite at the boundary after processing by our fourth-order partial differential equations.Experiments not only show that previous models of geodesic active contours are improved, allowing stable and quick boundary detection, but also region-based active contours can be less disturbed by noise.