基于相关向量机模型的能见度测量  

Visibility Measurement Based on Correlation Vector Machine Model

作　　者：陈强 叶青[1] CHEN Qiang;YE Qing(College of Electrical and Information Engineering,Changsha University of Science and Technology,Changsha 410000 China)

机构地区：[1]长沙理工大学电气与信息工程学院,湖南长沙410000

出　　处：《自动化技术与应用》2020年第3期86-90,97,共6页Techniques of Automation and Applications

摘　　要：基于支持向量回归机的能见度测量模型虽然克服了已有基于图像测量能见度方法的局限性,提高了测量的灵活性,但是模型中参数较多,模型优化难度大。为此,引入相关向量机构建模型用于能见度的测量,该模型参数较少更易实现模型优化。分别运用相关向量机以及支持向量回归机构建图像对比度特征与能见度值之间的关系模型,进行能见度值的测量,实验结果表明:相关向量机模型测量性能更佳。The visibility measurement model based on support vector regression overcomes the limitations of existing methods based on image measurement visibility and improves the flexibility of measurement. However, because there are many parameters in the model, it is very difficult to optimize the model further. Hence, a correlation vector machine model is introduced for visibility measurement, which is easier to be optimized with less parameter. The experiments adopted Support Vector Regression or correlation vector machine are compared. The experimental results show that the correlation vector machine model is better performance.

关 键 词：能见度测量 模型优化 支持向量回归机 相关向量机 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]