摘要
淋巴结的病理变化对检测恶性肿瘤和判断肿瘤转移是一个重要依据,因此淋巴结图像的质量对医生进行病理分析非常重要.文章将云模型与峰值阈值法相结合,自动获得模糊C均值(FCM)聚类的初始聚类中心,大大提升FCM的聚类效果,也较好地保留淋巴结图像的边缘细节,减轻离散频率分布图的离散度的影响.实验结果表明,文章的方法使图像中模糊目标区域的边缘更清晰.
The pathological change of lymph node is an important basis of malignant tumor detection and judg-ment of cancer metastasis. Therefore, the image quality of lymph node is very important for the doctors to con-duct pathological analysis. Based on the combination of cloud model and peak threshold method, the initial cluster centers of fuzzy C-means （FCM） clustering is obtained automatically, the clustering effect of FCM is en-hanced greatly, the image edge details of lymph node are also retained, the impact of the dispersion of a discrete frequency distribution is reduced. The experimental results show that the proposed method makes the image clearer in fuzzy edge of the target area.
出处
《广州大学学报（自然科学版）》
CAS
2013年第2期61-66,共6页
Journal of Guangzhou University:Natural Science Edition
基金
广东省自然科学基金资助项目(S2011040004121)
