基于特征加权和支持向量机的图像分类研究  

Image classification based on feature weighting and support vector machine

作　　者：杜娟[1] 孙君顶[1,2] 

机构地区：[1]河南理工大学计算机科学与技术学院,河南焦作454000 [2]"图像处理与图像通信"江苏省重点实验室,江苏南京210003

出　　处：《激光与红外》2013年第3期315-318,共4页Laser & Infrared

基　　金：教育部科学技术研究重点项目(No.210128);河南省骨干教师资助计划(No.2010GGJS-059);"图像处理与图像通信"江苏省重点实验室基金(No.LBEK2011002)资助

摘　　要：利用图像特征加权方法和支持向量机实现了图像的有效分类。首先根据特征的稳定性来判断特征的重要程度,从而赋予不同权重;然后借助支持向量机实现图像分类;最后采用不同颜色和纹理特征验证了在特征加权和不加权情况下图像分类的准确程度。实验结果表明本文的方法有效提高了图像分类的准确性。A new effective method for image classification is presented in the paper based on feature weighting and support vector machine. First of all, each dimension of the feature is set different weight according to its stability judged by the degree of its importance. And then, the images are classified based on the weighted feature and support vector machine. Finally, different color and texture features are adopted to verify the classification accuracy in the case of feature weighted and unweighted. Experimental resuhs show that the proposed method is effective to improve the ac- curacy of image classification.

关 键 词：底层特征 支持向量机 特征加权 图像分类 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]