Robustness-enhanced optical coverless hiding scheme for UAV images based on linear canonical transform and mapping...

Current optical image encryption and hiding methods generally suffer from weak robustness in the face of various attacks and transmission channel interference, which severely restricts their engineering application. In addition, existing embedded optical image hiding methods require modification of the carrier image, leading to limited embedding capacity and the risk of plaintext information leakage during the storage and transmission of the stego-carrier. Furthermore, there are few reports in the literature on optical encryption and hiding of UAV aerial images to date. To address these issues, this paper proposes what is believed to be a novel robust-enhanced optical coverless hiding method for UAV aerial images. In the proposed hiding method, bimodal biometric keys are first generated using face and palmprint images. Subsequently, the UAV aerial image is subjected to scrambling and zero-padding, followed by an encryption process performed sequentially through a cascaded Fresnel transform, a biometric mask, and a linear canonical transform. To achieve coverless hiding, a robust feature sequence is extracted from the carrier image, and the real and imaginary parts of the optically encrypted complex amplitude, after scrambling and diffusion, are subjected to XOR mapping to generate a mapping key, thereby realizing coverless hiding. Extensive numerical simulation results demonstrate that the proposed hiding method not only exhibits strong security and robustness against conventional attacks but also shows high reliability and robustness in practical communication systems such as OFDM and GFDM.