摘要
针对图像分类中的特征选择问题,提出一种多特征筛选与支持向量机相融合的图像分类模型.首先提取图像的多种特征,并对特征进行归一化处理;然后根据平均影响值对特征进行筛选,选择一组最优的特征子集;最后采用支持向量机构建图像的多分类器.采用图像数据集SIMPLIcity进行仿真实验验证该模型的有效性.实验结果表明,该模型降低了图像分类的开销,提高了图像分类性能.
Aiming at the feature selection problem in image classification, we proposed a multiple feature selection and support vector machine for an image classification model. Firstly, we extracted a varity of features of the image and normalized the features. Secondly, we selected a set of optimal feature subset according to the average impact value of features. Finally, the multi classifier was built by using support vector machine, and the simulation experiments were carried out to verify the validity of the model by using image data set SIMPLicity. The experimental results show that the proposed model can reduce the cost of image classification and improve the performance of image classification.
出处
《吉林大学学报（理学版）》
CAS
CSCD
北大核心
2016年第4期862-866,共5页
Journal of Jilin University:Science Edition
基金
河南省科技攻关项目(批准号:132102210423
122102210549)
关键词
图像处理
分类模型
多特征
支持向量机
image processing
classification model
multiple feature
support vector machine
