Traditional image encryption methods usually generate noisy ciphertext images, which are vulnerable to targeted analysis by attackers, thus exposing the existence of encrypted information. To solve this problem, this study proposes a visually meaningful asymmetric image encryption method based on the rotation transform domain and random devil’s vortex phase mask (RDVPM). The proposed method constructs RDVPM using devil lens function, Fresnel zone plate phase function and vortex phase function, and uses it as the public key of the encryption system. In addition, using visually meaningful images, amplitude and phase truncation techniques are used in the gyrator transform domain to construct an asymmetric encryption system to generate a specific Private key source, effectively improving the efficiency of encryption and decryption while ensuring that the system has a high degree of security. The visually meaningful image is a completely natural image that does not contain the original image information, which effectively improves the security of the encryption system. Numerical simulation results show that under the conditions of multiple test images, the proposed algorithm has high system parameter sensitivity and key sensitivity, and has good statistical characteristics. It has good robustness and security under Gaussian noise, speckle noise and salt and pepper noise attacks, occlusion attacks, chosen plaintext attacks and specific attack conditions. The proposed optical asymmetric encryption method provides a new idea for fast and secure visually meaningful image encryption. Recommended citation:

@article{liu2025visually, title={Visually meaningful asymmetric image encryption based on a random devil’s vortex phase mask in the gyrator transform domain}, author={Liu, Yujia and Sham, Chiu-Wing and Zhang, Fuqi and Zhao, Hui and Ma, Longyu}, journal={Optics and Lasers in Engineering}, volume={194}, pages={109121}, year={2025}, publisher={Elsevier} }