基于改进的双水平集的MRI图像快速分割方法

针对MRI图像具有高噪声与灰度不均的特点,提出了结合小波变换与中值滤波的去噪预处理的双水平集的快速分割方法。对于MRI图像存在的多种噪声问题,利用小波变换去除高斯噪声,采用中值滤波去除椒盐噪声,对原始图像进行预处理。在传统的双水平集模型中增加一个自适应的加速因子,对去噪图像进行快速分割得到分割效果图。实验结果表明,改进的算法显著加快了图像分割速度,既有较强的抗噪性,又保留了图像的细节信息,且无需重新初始化,取得了良好的分割效果。

Fast Segmentation Algorithm of MRI Image Based on Improved Dual Level Set

In view of the characteristics of high noise and uneven gray level in MRI image,a fast segmentation method of dual level set combining wavelet transform and median filter is proposed.For many kinds of noise problems in MRI image,wavelet transform is used to remove Gaussian noise,while median filter is used to remove salt and pepper noise,and the original image is preprocessed.An adaptive acceleration factor is added to the traditional dual level set model to segment the denoised image quickly and get the segmentation image.Experimental results show that the algorithm improves the speed of image segmentation,providing strong noise resistance and retaining the details of the image,unnecessary to be re-initialized,and achieves good segmentation results.