基于SVM的肺部图像分类  

Lung Image Classification Based on SVM

作　　者：张毅宇 ZHANG Yiyu(Xinjiang University,Urumqi Xinjiang 830000,China)

机构地区：[1]新疆大学,新疆乌鲁木齐830000

出　　处：《信息与电脑》2023年第17期60-63,共4页Information & Computer

摘　　要：针对肺部图片中肺部病症复杂,导致人工识别精度低的问题,提出基于支持向量机(Support Vector Machine,SVM)的肺部图像分类方式。首先,对图像进行预处理以提高图像质量,从图像中分离所需的肺部癌症对象。其次,通过灰度共生矩阵(Gray Level Coevolution Matrix,GLCM)提取6种肺部病症特征。最后,将提取的特征作为SVM的输入,利用SVM输出3种分类结果。实验结果表明,所提方法对于肺部病症分类的准确性高于90%。To address the problem of complex lung diseases in lung images,which leads to low accuracy of manual recognition,this paper proposed a lung image classification method based on Support Vector Machine(SVM).Firstly,preprocessing is done to improve the image quality and to separate the desired lung cancer objects from the image.Subsequently,extract the features of six lung conditions by Gray Level Coevolution Matrix(GLCM).Eventually,the extracted features are used as inputs to SVM,which is utilized to output the three classification results.The experimental results showed that the accuracy of lung condition classification was higher than 90%.

关 键 词：支持向量机(SVM) 肺部图像识别 灰度共生矩阵(GLCM) 

分 类 号：TP311.1[自动化与计算机技术—计算机软件与理论]