基于TWSVM的图像分类-南京师范大学学报自然科学版.PDFVIP

第37卷第3期 南京师大学报(自然科学版) Vol.37 No.3 2014年9月 JOURNAL OF NANJING NORMAL UNIVERSITY(Natural Science Edition) Sept,2014 詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨詨 基于 TWSVM 的图像分类 1 1,2 朱志宾 ,丁世飞 (1.中国矿业大学计算机科学与技术学院,江苏 徐州221116) (2.中国科学院计算技术研究所智能信息处理重点实验室,北京 100190) [摘要]摇 图像分类技术是图像数据处理中最重要的技术之一.支持向量机是基于统计学习理论而提出的机器 学习算法,在样本数少的时候能达到很好的分类效果.孪生支持向量机是基于支持向量机而提出来的,其性能优 于支持向量机.通过提取彩色图像的颜色特征与纹理特征,利用孪生支持向量机与支持向量机对这些特征向量 进行分类,孪生支持向量机的分类准确率与稳定性都高于支持向量机. [关键词]摇 图像分类,支持向量机,孪生支持向量机,特征提取 [中图分类号]TP181摇 [文献标志码]A摇 [文章编号]1001-4616(2014)03-0008-07 Image Classification Based on Twin Support Vector Machines 1 1,2 ZhuZhibin ,Ding Shifei (1.School of Computer Science and Technology,China University of Mining and Technology,Xuzhou221116,China) (2.Key Laboratory of Intelligent Information Processing,Institute of Computing Technology,Chinese Academy of Sciences,Beijing 100190,China) Abstract:The image classification is one of the most important technologies of image data processing.Support Vector Machine is a machine learning algorithm based on statistical learning theory,which can achieve a good classification results in the small sample size.Twin Support Vector Machinesisbasedon Support Vector Machine,whichissuperiorto Support Vector Machine.By extracting colorfeaturesandtexturefeaturesof images,usingTwin Support Vector Machines and Support Vector Machine to classify these feature vectors,the results shows that the accuracy and stability of Twin Support Vector Machines is higher than Support Vector Machine. Key word