An image encryption scheme based on chaotic system and compressed sensing for multiple application scenarios *

In this paper, we propose a novel image encryption scheme based on a chaotic system and compressed sensing for multiple application scenarios. It is a dedicated scheme for JPEG format that makes full use of the characteristics of DCT data. The proposed scheme generates a key from a plain image with random noise, which is then used to generate the parameters needed in the scheme through using a new chaotic system structure. This structure employs data segmentation for the DCT data of a plain image and processes sparse data and non-sparse data separately: it permutes non-sparse data, and applies permutation and compressed sensing to sparse data. By merging the processed data and performing efficient confusion, we obtain the final encrypted data. The reverse operation is image decryption, and partial decryption can be performed quickly since sparse data is not required. Such partial decryption can be used for multiple applications, such as copyright protection platforms and machine learning based on private data. Simulation experiments and a performance analysis show that the key generation method and efficient confusion of our scheme can improve the security of image encryption. Furthermore, the data segmentation and reconstruction algorithm based on a cyclic matrix can greatly improve decryption efficiency.