顺序统计滤波能量驱动的鲁棒快速图像分割

为了提升图像分割的速度与初始轮廓和参数鲁棒性，提出了一种基于有序统计滤波能量驱动的鲁棒主动轮廓模型。首先用顺序统计滤波的边力函数来代替传统的数据拟合，进一步引入边力函数能量泛函快速自适应地吸引曲线向目标的边界演化。为了有效地调节水平集函数，利用优化后的长度项对曲线进行平滑和缩短，并通过标准梯度下降法将能量泛函最小化。最后实验结果表明提出的方法能够更加快速、准确地分割图像，并且对初始轮廓和参数具有较强的鲁棒性。

Robust Fast Image Segmentation Driven Based on Sequential Statistical Filtering Energy

In order to improve the speed of image segmentation and the robustness of initial contour and parameters, a robust active contour model based on the energy driven of ordered statistical filtering was proposed. First, the edge force function of sequential statistical filter was used to replace the traditional data fitting, and then the energy functional of edge force function was introduced to attract the curve to the boundary of the target. In order to adjust the level set function effectively, the optimized length term was used to smooth and shorten the curve, and the standard gradient descent method was used to minimize the energy functional. The experimental results show that the proposed method can segment the image more quickly and accurately, and has strong robustness to the initial contour and parameters.