针对医学图像在采集传输等过程中易受噪声污染,且目前多数去噪方法对混合噪声去噪效果不好,影响三维重构精度的问题,提出了基于自适应三维分数阶积分和中值滤波结合的混合噪声去除方法.首先介绍了几种传统方法在去除混合噪声中的不足,然后基于三维图像梯度信息提出了三维分数阶积分的自适应分数阶阶数,利用分数阶积分和中值滤波的各自优点,将两者结合对混合噪声去噪,并提出了基于混合去噪的边缘曲面追踪算法.实验结果和数据分析表明,提出的混合去噪方法能够从噪声污染的医学图像切片中追踪出高精度边缘曲面,与传统去噪方法相比,具有更好的去噪效果.

Medical images are usually polluted by noise during the process of transmission and collection.A mixed noise removing method based on adaptive three-dimensional (3D) fractional integral and median filtering was proposed to reconstruct high accuracy edge surfaces.Firstly,some traditional methods in denoising were introduced.Then the adaptive order of fractional integral based on the gradient information of three-dimensional image was proposed.Making use of the advantage of fractional integral and median filtering,an edge surface tracking algorithm based on mixed noise denoising method was proposed.The experimental results show that the proposed method can effectively track edge surfaces from noisy medical image slices with high accuracy.The denoised results are better than traditional denoising methods.