基于粗糙集和支持向量机的肿瘤图像识别  

A New Method for Tumor Images Recognition Based on Rough Sets and SVMs

作　　者：张海燕[1] 蓝集明[2] 

机构地区：[1]四川理工学院理学院,四川自贡643000 [2]四川理工学院计算机学院,四川自贡643000

出　　处：《四川理工学院学报（自然科学版）》2014年第6期38-42,共5页Journal of Sichuan University of Science & Engineering(Natural Science Edition)

基　　金：四川省教育厅科研项目(14ZB0214);自贡市科技局科研项目(2013ZC11);企业信息化与物联网测控技术四川省高校重点实验室项目(2014WYJ03);桥梁无损检测与工程计算四川省重点实验室科研项目(2014QZY01)

摘　　要：设计了一种基于粗糙集与支持向量基的乳腺肿瘤图像识别方法。其基本思想是首先对图片进行降噪预处理,接着提取纹理和形状特征构成表征医学图像的特征矢量,然后将特征离散归一化处理,再用粗集方法进行特征属性约简,最后利用支持向量基进行识别。结果表明,该方法取得了比较理想的识别效果。In order to describe the output characteristics of a vertical axis wind turbine, the wind speed model, drivesystem and speed control system of a vertical-axis wind turbine are established in MATLAB/SIMULINK. Particularly, takinga 5KW vertical axis wind turbine as example, the power coefficients under the different tip speed ratio are received by usingDouble-Multiple Streamtube （DMS） model. Then the data is imported into MATLAB, and the mathematical model of the5 KW vertical axis wind turbine rotor is obtained by using the curve fitting toolbox, thus the SIMULINK model is built. Laterthe wind turbine model is formed by these above models and other models. Some operation parameters of wind turbine aregiven, and the data for analysis is received. Through the simulation study, MATLAB/SIMULINK can better simulate theoverall performance of a wind turbine from wind turbines to the drive system, which provides a reference for the futureresearch of the overall performance of vertical axis wind turbines.

关 键 词：医学图像 肿瘤识别 粗糙集 支持向量机 特征提取 属性约简 

分 类 号：TP31[自动化与计算机技术—计算机软件与理论]