Image classification method and system based on auto-encoder and decision tree

The invention discloses an image classification method and system based on an auto-encoder and a decision tree, and the method comprises the steps: collecting image sample data, and converting an image sample into a pixel information matrix/vector; learning representation information of the image samples by using an auto-encoder network model, and compressing and extracting low-dimensional feature information of the image samples by using an encoder; updating the nearest neighbor value corresponding to each sample in the process of iteratively solving the optimal weight parameter of the auto-encoder network; based on low-dimensional sample feature information extracted from the trained auto-encoder network model, combining a sample nearest neighbor value obtained by iteration as a corresponding sample label to construct a decision tree model; and obtaining low-dimensional feature information of a new sample by using an auto-encoder, inputting the low-dimensional feature information into the decision tree to o