改进的FSVM结合语义特征的甲状腺图像分类方法  

Thyroid Image Classification Method of Combining Improved FSVM with Semantic Features

作　　者：赵杰[1] 万丹丹[1] 门国尊[2] 

机构地区：[1]河北大学电子信息工程学院,河北保定071000 [2]河北大学经济学院,河北保定071000

出　　处：《电视技术》2014年第5期20-23,共4页Video Engineering

基　　金：河北省卫生厅科研基金项目(20120395)

摘　　要：由于甲状腺发病率高且图像良恶性难以分辨,提出一种改进模糊支持向量机(FSVM)结合语义特征的甲状腺图像分类方法。通过概率潜在语义分析(PLSA)模型对给定的图像训练样本提取语义特征,输入到FSVM中进行分类。其中隶属度是影响FSVM分类精确性的关键,故对其进行改进,在考虑样本点到类中心距离的基础上,对样本点间的紧密度也进行了估计。利用训练生成不同的FSVM测试图像,采用集成方法将分类结果集成,避免了单分类器的分类误差。实验结果表明,该方法可获得较好的分类结果。Because the high incidence of thyroid and it is difficult to distinguish the benign and malignant of a image, so putting forward an thyroid im-age classification method combined improved fuzzy support vector machine（FSVM）with semantic features. Extracting semantic features for given training images samples through the probabilistic latent semantic analysis（PISA） model then classified by FSVM. Improved the membership degree for it is the key to affect the classification accuracy of FSVM,tightness among the sample points are also estimated on the basis of the distance from a sample point to the class center. Using training to generate different test image for FSVM,avoid the error by single classifier,using classifiers integrated method to in-tegrate the classification results. The experimental results show that this method can achieve better classification results.

关 键 词：语义特征 模糊支持向量机 隶属度 甲状腺图像 集成分类 

分 类 号：TN919.8[电子电信—通信与信息系统]