非负矩阵分解(NMF)能够提取图像的局部特征,是一种基于局部的数据挖掘方法,在一定程度上勾勒出了相关图像在基矩阵所代表空间上的分布,但NMF并未考虑数据的内在几何结构.提出了一种新颖的基于非负矩阵分解与非线性降维方法Isomap相结合的新方法,全局的非线性降维方法Isomap能发现数据的内在结构和相关性,使高维数据在低维空间变得可视化.将本算法应用于图像检索,实验表明,该方法能够更加准确地获取信息,提高检索的准确性.%The non-negative matrix factorization(NMF) is a local data mining method which can extract the local feature of a image,and it can describe the relevant image distribution on the space of base matrix.But NMF neglects the inner geometric structure of data.This paper proposed a new method which integrated NMF and non-linear dimensionality reduction Isomap.The global dimensionality reduction method could discover the inner structure and relativity of the data, it made the high dimension data visualization on lower space.In image retrieval experiment, the method can obtain information more precise and improve the accuracy of retrieval.
展开▼