摘要
支持向量机方法被看作是对传统学习分类方法的一个好的替代,特别在小训练样本、高维情况下,具有较好的泛化性能。该文采用了支持向量机方法对多目标图像进行了分割研究。实验结果表明:模型参数对支持向量机方法的分割性能有较大的影响;对多目标图像的分割,支持向量机方法是一种很有前景的分割技术。
Support Vector Machine approach is considered a good candidate because of its good generalization performance,especially when the number of training samples is very small and the dimension of feature space is very high.The presented paper investigates the segmentation of multi-target image based on Support Vector Machine approach.Experimental results show that:the influence of model parameters on the segmentation performance of Support Vector Machine approach is significant;Support Vector Machine approach is a promising technique for the segmentation of multi-target image.
出处
《计算机工程与应用》
CSCD
北大核心
2005年第15期11-12,137,共3页
Computer Engineering and Applications
基金
国家自然科学基金资助(编号:60475024)
关键词
多目标图像分割
支持向量机
统计学习理论
segmentation of multi-target image,Support Vector Machine,Statistical Learning Theory
