摘要
为提高MRI、PET等医学图像分割的精度和速度,提出了基于核聚类的MRI和PET医学图像分割方法,通 过利用Mercer核,将原本简单的样本特征映射到更复杂的高维空间中去,放大了样本特征间的差异,这样能快速 准确地分割样本。实验表明,基于核聚类的分割方法在医学图像处理中具有重要的应用价值。
In order to improve the precision and speed of medical image segmentation of MRI and PET, we present a method based on the kernel clustering. By using Mercer kernel functions, we can map the data in the original simple space to a high-dimensional complicated space in which we can amplify the feature differences of the data and finish the segmentation with fast and accurately. Our experiment proves this method to be important in medical image segmentation.
出处
《宿州学院学报》
2005年第1期88-90,共3页
Journal of Suzhou University
关键词
聚类分析
核函数
医学图像分割
Clustering analysis
kernel function
medical image segmentation
