在分析带标记线的心脏MRI特点的基础上,提出了利用Gabor滤波与Chan和Vese模型相结合对带标记线的心脏MRI图像进行分割的方法。通过对带标记线的心脏MRI图像进行Gabor变换,把标记线分离出来,并且扩大心肌和血液之间的差异,再对变换后的图像用Chan和Vese模型进行分割。实验结果表明了该方法的有效性。
Based on the characteristics of tagged cardiac MRI, this paper presented a new method for segmenting tagged cardiac MRI image,which combined Gabor filter banks and Chan-Vese model. The purpose of Gabor filter banks was to suppress the tags in the myocardium, and enhance the blood-to-myocardium contrast. It segmented the filtered image by using Chan-Vese model. The results of the experiments have proved its effectiveness for segmenting tagged cardiac MRI image.