基于递增权函数的邻接矩阵与非负矩阵分解的图像分类方法

Image classfication using increasing weighting function of adjacency matrix of graph and non-negetive matrix factorization

将递增权函数的邻接矩阵和非负矩阵分解方法相结合,应用于图像分类.首先由图像中提取的特征点构造递增权函数的邻接矩阵,再对其进行非负矩阵分解,用分解后的特征向量作为PNN分类器的输入,实现对图像的分类.算法的可行性和准确性通过模拟图像和真实图像的多组实验得到了验证.

In this paper,the adjacency matrix of graph based on the increasing weighting function combined with the method of non-negative matrix factorization was applied to the image classification.First,the character points could be distilled from different images.Then,these points were used to construct the adjacency matrix of the increasing weighting function,and the eigenvector of the image could be obtained by the non-negative factorization of the adjacency matrix.Finally,the eigenvector was put into PNN（Probabilistic Neural Network）classifier to accomplish the image classification.Several groups of experiments were presented between simulating images and real images.The results showed that the method presented in this paper was feasible and accurate.