摘要
针对现有的脑肿瘤模糊目标边缘分割方法,存在分割准确率低的问题,提出核磁共振图像的脑肿瘤模糊目标边缘分割方法。依据词袋模型构建视觉词典,采用密集采样方法选取脑肿瘤核磁共振图像,根据P-M模型与最大值-最小值方法去噪以及灰度化处理图像,将处理后的图像依据滑动窗口提取像素点特征,构建图像t混合模型。采用EM算法求解脑肿瘤模糊目标边缘,实现了脑肿瘤模糊目标边缘的分割。仿真对比实验结果显示,相较于现有脑肿瘤模糊目标边缘分割方法来看,提出的脑肿瘤模糊目标边缘分割方法极大的提升了分割准确率,充分显示提出的脑肿瘤模糊目标边缘分割方法具备更好的分割效果。
Currently, the fuzzy target edge segmentation method of cerebral tumor has the problem of low accuracy. Therefore, a method to segment the fuzzy target edge of brain tumor based on MRI image was put forward. According to the bag of words model, a visual dictionary was constructed, and then the MRI image was selected by the intensive sampling method. Moreover, the P-M model and the maximum minimum method were used to remove the noise from image and grayed the image. According to the sliding window, the pixel features were extracted from the processed image, and then the hybrid model of image t was constructed. Finally, EM algorithm was used to solve the fuzzy target edge of cerebral tumor. Thus, the segmentation of fuzzy target edge was realized. Simulation results show that, compared with the existing methods, the proposed method greatly improves the segmentation accuracy, which fully shows better segmentation effect.
作者
李依桐
陈悦
杨皙睿
LI Yi-tong;CHEN Yue;YANG Xi-rui(School of Information and Electronics,Beijing Institute of Technology,Beijing 100081,China)
出处
《计算机仿真》
北大核心
2020年第10期369-373,共5页
Computer Simulation
