Polarized laser and image fusion material identification method of sparse auto-encoder

The invention discloses a polarization laser and image fusion material identification method for a sparse auto-encoder, and the method comprises the steps: inputting a to-be-identified object sample into a constructed prediction classification model, and obtaining the material type of the object sample. Constructing a prediction classification model: constructing a prediction classification initial model based on a framework composed of a sparse auto-encoder and a softmax classifier; training a prediction classification initial model by using the obtained training set to obtain a prediction classification model; and testing the prediction classification model by using the obtained test set, and if a test condition is met, indicating that the prediction classification model is qualified. According to the method, the training set is used for training the frame formed by the sparse auto-encoder and the softmax classifier, and the precision of material recognition is improved. 本发明公开了一种稀疏自编码器的偏振激光和图像融合的材质识别方法，将待识别