论文部分内容阅读
给出了一种基于水平集演化、无需任何先验信息的SAR图像分割方法.该方法是一种基于区域信息的统计活动轮廓模型方法,通过利用分段阶跃函数估计图像概率密度函数,克服了利用特定概率分布模型估计概率密度函数时,需要利用先验信息预先假定图像概率分布模型的问题;通过引入惩罚项,避免了费时且难于操作的水平集函数重新初始化过程.还给出了具体的数值实现方案和相关参数取值,改进了数值实现中的迭代终止条件.实验结果表明,固定使用列出的参数,无需任何人为干预,对于大多数图像都可获得令人满意的分割结果;对于少数图像,通过简单的参数调整也可得到良好结果.
A SAR image segmentation method based on level set evolution without any priori information is presented. This method is a statistical active contour model based on region information. By using the step-by-step function to estimate the image probability density function, When the probability density function is estimated by a specific probability distribution model, the problem of the image probability distribution model needs to be pre-assumed by using the prior information. By introducing the penalty items, the time-consuming and difficult-to-operate level set function reinitialization process is avoided. The iterative termination conditions in the numerical implementation are improved.The experimental results show that the parameters listed are fixed without any human intervention and satisfactory segmentation results can be obtained for most images.For the minority Image, with simple parameter adjustments can also get good results.