基于局部区域信息的医学图像分割及偏移场矫正方法

Segmentation and bias corrected method for medical image based on local region information

针对医学图像中由于偏移场的存在而导致图像灰度不均匀的问题,提出了一种基于局部区域信息的医学图像分割及偏移场矫正方法,以矫正偏移场使图像变为灰度均匀。该方法利用图像局部区域信息,通过拟合图像和原始图像构造能量函数,采用变分水平集方法进行求解。实验结果表明,该方法能够有效地实现医学图像分割及偏移场矫正,与其他分割及偏移场矫正方法相比,该方法具有较高的分割及偏移场矫正的精度和效率。

To address the bias field leading to image intensity inhomogeneity of medical image,this paper proposed a segmentation and bias corrected method for medical image based on local region information,to make the image intensity homogeneity. Based on the local region information of image,this method constructed the energy function by fitting image and original image,which was minimized by variational level set method. Experiment results show that the presented method can accurately and fast segment and correct the medical image. Compared with the traditional methods,this method has more precise and efficient in segmentation and bias corrected.