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

This article presents a novel image compression and encryption framework based on a newly proposed chaotic map called the modSine map, which exhibits high sensitivity to initial parameters and enhanced chaotic behavior compared to existing one-dimensional maps. The method integrates compressive sensing (CS) for simultaneous compression and encryption, designing the measurement matrix from the modSine map and employing a user name-based encryption scheme that generates encryption keys from statistical features of the authorized user’s name. Validation through simulations on multiple standard images demonstrates effective image recovery with peak signal-to-noise ratio (PSNR) around 32 dB and structural similarity index (SSIM) near 0.86 at a sampling ratio of 0.5, alongside robustness against noise, occlusion, and key sensitivity attacks. The approach achieves a large key space (approximately 2^296), ensuring resistance to brute force attacks, and compares favorably in reconstruction quality and security metrics with existing CS-based image encryption methods.