首页 | 官方网站   微博 | 高级检索  
     

遗传算法优化特征权值的支持向量机图像分类
引用本文:王东霞,周观民.遗传算法优化特征权值的支持向量机图像分类[J].电视技术,2015,39(2).
作者姓名:王东霞  周观民
作者单位:济源职业技术学院信息工程系,河南济源,459000
基金项目:河南省科技攻关项目(132102210229)
摘    要:为了提高图像分类的准确率,提出了一种遗传算法优化特征权值的支持向量机图像分类(GA-SVM).首先分别提取图像的颜色和纹理特征,然后采用改进遗传算法确定特征权值,最后采用支持向量机建立图像分类器,并对corel图像库进行仿真测试.结果表明,相对于其他图像分类算法,GA-SVM提高了图像分类精度.

关 键 词:图像分类  特征加权  支持向量机  遗传算法
收稿时间:2014/4/21 0:00:00
修稿时间:6/9/2014 12:00:00 AM

IMAGES CLASSIFICATION BASED ON GENETIC ALGORITHM OPTIMIZING FEATURES WEIGHT AND SUPPORT VECTOR MACHINE
wang,dong xia and zhou guan min.IMAGES CLASSIFICATION BASED ON GENETIC ALGORITHM OPTIMIZING FEATURES WEIGHT AND SUPPORT VECTOR MACHINE[J].Tv Engineering,2015,39(2).
Authors:wang  dong xia and zhou guan min
Affiliation:Jiyuan Vocational and Technical College,Jiyuan Vocational and Technical College
Abstract:Abstract In order to improve the accuracy of the image classification, this paper proposes a support vector machine image classification (GA-SVM) based on genetic algorithm optimizing feature weight. Firstly, the color and texture of image are extracted, then the feature weight are determined by the improved genetic algorithm, finally, image classifier is established by support vector machine and the simulation test is carried out on core image library. The results show that the GA-SVM algorithm can improve the image classification accuracy compared to other image classification algorithm.
Keywords:Keywords image classify  feature-weighted  support vector machine  genetic algorithm
本文献已被 万方数据 等数据库收录!
点击此处可从《电视技术》浏览原始摘要信息
点击此处可从《电视技术》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司    京ICP备09084417号-23

京公网安备 11010802026262号