User Name-Based Compression and Encryption of Images Using Chaotic Compressive Sensing Theory.

Simultaneous compression and encryption of images using a novel chaotic map is proposed in this paper. Both compression and encryption of images are carried out by the theory of compressive sensing (CS). A novel chaotic map with a high degree of chaos that is extremely sensitive to its initial parameters is proposed. A measurement matrix for the CS framework is designed based on the proposed map. The compression and recovery of images with different compression/sampling ratios are tested using the designed measurement matrix. Encryption of the compressed data is carried out using the proposed chaotic map and a novel user name-based encryption scheme. The entire encryption/decryption process proposed is completely dependent on the sequence obtained from the proposed chaotic map as well as the authorized user name. Thus, by this process, only authorized people with a valid user name will be able to decrypt the encrypted data and recover the actual underlying image. Simulation results on the proposed scheme with different images show that the average peak signal-to-noise ratio and structural similarity index values of about 32 dB and 0.861 are obtained for a sampling ratio of 0.5. Validations on the proposed map and the encryption process that were carried out using various standard tests prove the efficiency of the system in successfully compressing and encrypting the images. Also, the qualitative evaluation of the proposed compression–encryption process outperforms some of the existing works in the literature. [ABSTRACT FROM AUTHOR]