Rapid multi-view semi-supervised learning method and system based on learning graph

The invention discloses a fast multi-view semi-supervised learning method and system based on a learning graph, and belongs to the field of multi-view image classification. According to the fast multi-view semi-supervised learning method based on the learning graph, a multi-view image classification task can be effectively completed on the premise that the calculation complexity of a multi-view image classification method based on the graph is reduced. Different from most existing methods for constructing an n * n traditional similarity graph, m (m is more than or equal to n) representative anchor point images are selected for n images, and the n * m anchor point graph is constructed to accelerate multi-view image classification work. The weights are automatically distributed for the anchor point diagrams with different image features, and the optimal multi-view consistency anchor point diagram is obtained through joint learning of the original image features and the prediction labels. 本发明公开了一种基于学习图的快速多视图半监督学