为提高计算机辅助诊断的准确性，提出一种基于小波变换和改进的模糊C均值（ Fuzzy C－Means， FCM）算法的医学CT图像分割方法。以FCM算法为基础，首先利用小波变换对医学图像进行分解，用分解后低频图像的像素点作为FCM算法的样本点；其次，利用马氏距离来进一步修正 FCM＿S（ FCM＿Spatial）算法，修正后的 FCM 算法能更加精确地反映医学图像的信息。实验结果表明，算法的效率得到较大提高。%In order to enhance the accuracy of computer auxiliary diagnosis, a medical CT image segmentation algorithm based on wavelet transform and improved FCM algorithm is proposed .Because the traditional FCM algorithm usually run on all im-age pixels, which makes the efficiency of the algorithm reduced.On the basis of FCM algorithm, firstly this algorithm processes the image using wavelet transform, and the low frequency images by wavelet transform are inputted into FCM algorithm to obtain seg-mentation results.It not only greatly reduces the time complexity of the algorithm but also effectively suppresses image noise .Sec-ondly, the algorithm introduces the Mahalanobis distance to improve FCM_S algorithm, and the improved FCM algorithm can be more accurate to obtain medical image information .The experiments show that this algorithm significantly improves the segmenta-tion’s efficiency.
展开▼
