摘要
针对当前支持向量机支持优化的参数无法获得高精度的体育视频分类结果的难题,为了提高体育视频分类正确率,提出基于蚁群优化算法优化支持向量机的体育视频分类方法。首先采集体育视频,并提取体育视频分类的多个特征;然后采用主成分分析算法对体育视频分类特征进行处理,作为支持向量机的输入,体育视频类别作为支持向量机的输出,建立体育视频分类模型,并采用蚁群优化算法对支持向量机进行优化;最后采用多个体育视频数据进行分类仿真实验,结果表明,蚁群优化算法优化支持向量机的体育视频分类正确率高于90%,降低了体育视频分类错误,体育视频分类效果明显优于当前其他类型的体育视频分类方法,而且体育视频分类效率得到有效的改善。
In order to solve the problem that the current support vector machine(SVM)can′t support optimized parameters to obtain high⁃precision sports video classification results,the sports video classification method based SVM optimized by ant colony optimization(ACO)is proposed to improve the accuracy of sports video classification.The sports videos are collected and several features of sports video classification are extracted firstly,and then the classification features of sports video are processed by means of principal component analysis algorithm.The sports video classification features are taken as the input of SVM and the sports video category as the output of SVM to establish a sports video classification model.The SVM is optimized by ACO.A classification simulation experiment is conducted with multiple sports video data.The simulation experiment results show that the classification accuracy of the sports video classification based on SVM optimized by ACO is higher than 90%,and the classification error of sports video classification is reduced.The classification effect of sports videos is obviously better than that of other current sports video classification methods,and the classification efficiency of sports videos is effectively improved.
作者
王杨
刘蒙
闫伟光
WANG Yang;LIU Meng;YAN Weiguang(Zhangjiakou Vocational and Technical College,Zhangjiakou 075000,China)
出处
《现代电子技术》
北大核心
2020年第1期56-58,62,共4页
Modern Electronics Technique
