Intelligent model watermark embedding and extracting method and system based on convolutional neural network

The invention relates to an intelligent model watermark embedding and extracting method and system based on a convolutional neural network. The watermark embedding process of the method comprises thesteps that a data set containing a public data set and a non-public data set is established, data in the public data set only comprises key features, and data in the non-public data set comprises keyfeatures and implicit non-key features; data feature extraction is performed on the data set; an intelligent model is trained based on a convolutional neural network, and during model training, overfitting training is performed on a decision boundary implicit with non-key features, so as to embed a watermark into the intelligent model; the watermark extraction process comprises the steps that a non-public data set serves as a secret key, the secret key serves as input data, a legal copy intelligent model is called through an API on a server, a returned query result is compared with a predictedclassification result, and