相关向量机在车型识别中的应用研究

Research on application of relevance vector machine in car model identification

相关向量机是一种稀疏的贝叶斯学习算法,对非线性、高维数的小样本问题有非常好的分类效果和学习推广能力。而且使用较少的核函数,研究了用相关向量机技术进行车型识别,设计了基于相关向量机的车型分类器。实验结果表明,基于相关向量机的车型分类器不仅具有基于支持向量机的车型分类器的相同性能,而且比支持向量机使用更少的核函数,实验取得了较好的分类效果。

Relevance vector machine is a sparse Bayesian learning algorithm,it has good ability of classification and generalization for the nonlinearity,multi-dimension,and small-sample problems.A car model identification based on RVM is studied.Experimental results show that RVM classifier achieves comparable recognition accuracy to the SVM classifier,yet provides a full predictive distribu-tion,and also requires substantially fewer kernel functions.Experimental results are efficient for classification.