以医学图像为研究对象,针对任何一类特征都不能很好地表达医学图像的缺点以及进一步提高医学图像的识别率,提出了一种基于特征级数据融合与决策级数据融合相结合的分类方法。实验结果表明,采用特征级数据融合,融合后的特征可以较好地表达医学图像,且减少了后期分类的计算量;采用决策级数据融合,取得了比单个分类器更高的识别率。
This paper used the medical images as the study object.As the shortcomings of the features of any class can not express the medical images efficiently and in order to improve the recognition rate of the medical images further,proposed a new algorithm.This algorithm was a classification algorithm that combined the feature-level data fusion with the decision-level data fusion.As the results show,using the feature-level data fusion,the fused features can express the medical images better and reduce the computation of the later classification,and using the decision-level data fusion, a higher recognition rate can be got compared with single classifier.