Image processing method and device based on neural network and storage medium

Abstract

The invention relates to an image processing method and device based on a neural network and a storage medium, and the method comprises the steps: obtaining encryption noise according to a first image, a preset secret key and a noise generator; obtaining a second image according to the first image and the encryption noise; and obtaining a related prediction result of image encryption through the second image and the encrypted neural network. In the embodiment of the invention, as the encrypted neural network is only effective for the image encrypted by the preset key, the encrypted neural network cannot be directly applied to the original image under the condition that the structure and parameters of the neural network are stolen, thereby improving the security of image processing; meanwhile, because the encrypted image still maintains the visual effect of the original image, after the original image is encrypted and then distributed to equipment carrying the encrypted neural network,the encrypted neural network can be used for prediction, and the correctness of the prediction result can be further verified according to the encrypted image.

Description

technical field [0001] The present disclosure relates to the technical field of computer image processing, in particular to an image processing method, device and storage medium based on a neural network. Background technique [0002] Artificial Intelligence (AI) technology represented by neural networks has been applied to many scenarios such as medical diagnosis, face recognition, and automatic driving. In order to improve the security of image processing, the neural network model can be encrypted, so that even if the neural network model is stolen, it cannot be used directly. [0003] In related technologies, the original data is first encrypted, and then the encrypted data is used for training or prediction of the neural network model. However, the encryption process in this scheme will change the visual effect of the image itself. At the same time, since data encryption and neural network prediction are integrated, if the encrypted data is distributed to other devices ...

AI Technical Summary

Problems solved by technology

However, the encryption process in this scheme will change the visual effect of the image itself. At the same time, since data encryption and neural network prediction are integrated, if the encrypted data is distributed to other devices for prediction, the image received by the device will no longer have the original visual function, unable to verify and review whether the prediction results are correct

Images