水平集方法中速度停止项的改进

针对传统水平集算法只对含有加性噪声的图像有较好处理结果的缺点,对水平集函数的速度停止项进行了改进。首先,改进后的速度停止项可以通过自适应滤波来消除乘性噪声对水平集函数演化的影响;其次,提出了一种估计图像平滑区域的方法,以使水平集可以更准确的分割出图像中的目标;最后,再将该速度停止项带入到水平集演化方程中,对图像进行演化运算。实验结果表明,该方法能够在抑制乘性噪声的同时较好的对图像进行分割。

A Modified Stop Criterion in Level Set Method

A moditied speed stop item in level set method was proposed in order to solve the problem of the traditional level set that can only deal with the image with additive noise. In order to make the evolving front insensitive to the multiplicative noise,a new speed stop item based on adaptive filter is presented. And the proposed adaptive filter is used to obtain a local contrast image that is a parameter of speed stop item. Then a new method that evaluates the variance of homogeneous area is proposed. Last using the decreasing function of the local contrast image as the stop item evolves the level set function then gets the final segmentation result. The experimental results showed that the developed algorithm could segment the image well while suppressing the multiplicative noise.