To address the protection and secure transmission of digital orthophoto map (DOM) information in network environments, this study proposes a novel DOM encryption method that combines QR code and GS phase retrieval in two stages: ① The image data (RGB channels) of the DOM is encoded using the GS algorithm to generate three phase-only holograms. These holograms are scrambled and diffused using pseudorandom sequences generated by the Lorenz hyperchaotic system, enhancing the sensitivity of the optical keys and the ciphertext's sensitivity to plaintext changes through chaotic system properties and diffusion mechanisms. ② The geographic information of the DOM, along with partial keys and user authorization information, are encoded into a QR code and embedded into the three phase-only holograms. The three holograms with embedded QR codes are then synthesized into a meaningless white noise color image, serving as the final ciphertext. Experimental results for DOM in different scenes demonstrate that this method achieves unified encryption of both the DOM image data and geographic information. The decrypted images are highly consistent with the original plaintext images, achieving a peak signal-to-noise ratio of over 39 dB. Evaluation metrics including the correlation coefficient, structural similarity index measure, change rate of pixel number, and normalized change intensity, are close to their ideal values, outperforming existing methods. In addition to traditional chaos parameters, the GS parameters, such as diffraction distance and reference light wavelength, serve as the key, ensuring the method has a large key space (approximately 10141). Minor changes to the plaintext completely alter the ciphertext, and pixel values are statistically independent. The method exhibits strong robustness against ciphertext-only attacks, known-plaintext attacks, chosen-plaintext attacks, chosen-ciphertext attacks, and noise attacks, making it highly valuable for the protection and secure transmission of raster geographic data, including DOM.
