<p>The traditional approach of converting secret digital images into cipher images draws attention of third parties regarding the presence of some secret data behind the encrypted images. To prevent detection by the attackers, a visually meaningful encryption (VME) technique converts secret digital images into noise-like patterns. It conceals these pattern within the carrier data using steganographic methods, thereby avoiding the attention drawn by third parties to traditional encrypted images. In this paper, we introduce a novel and robust authentication-based visually meaningful image encryption (VMIE) scheme, achieved through the integration of a public key-based digital signature and encryption using chaotic maps, along with a lifting wavelet transformation (LWT) enabled embedding scheme. The scheme first achieves partial encryption using the public key cryptosystem and the 3D logistic map. The resulting ciphered image is then used to generate a digital signature, which is encrypted further by using the MRMAC. Finally, the chaotic key sequences from 3D logistic map employed to complete the encryption, followed by embedding into a carrier image using LWT. The novelty of the proposed scheme lies in the use of digital signatures, which not only verify integrity but also enhance the quality of encryption. Embedding through shuffling in the sub-band increases the security level compared to direct embedding. All the obtained VME images achieve strong imperceptibility and concealment for the encrypted data. The proposed technique’s robustness, efficiency and secrecy are verified through the security and statistical analysis. Additionally, comparison analysis demonstrates that efficiency and robustness become superior to some recent techniques. Additionally, it offers a high-quality reconstruction capability, while ensuring the quality of secrecy in visual encryption and facilitating proper decryption without data loss.</p>