Considering the necessity of identity authentication in most scenarios, a dual-channel (digital channel and optical channel) image encryption and authentication algorithm with computational ghost imaging (CGI) is proposed. The entire encryption process is governed by the pseudo-random sequences obtained by the 4D chaotic system and the Logistic map. During the digital channel encryption, the confusion and diffusion of the plaintext image are achieved by random scrambling and DNA computing, respectively. To alleviate the burden of key transmission and storage in the CGI encryption system, the speckle measurement matrices are constructed with the pseudo-random numbers obtained by the Logistic map. Additionally, a nonlinear encryption operation is carried out to compensate the linear vulnerability of CGI encryption. In the optical encryption channel, the authentication image is first encrypted with the CGI method, followed by the secondary encryption of the intensity values recorded by a bucket detector to produce an intermediate ciphertext. The final encryption image is generated by merging the intermediate ciphertexts from both encryption channels. It is demonstrated that the proposed optical image encryption and authentication scheme with CGI provides satisfactory encryption and decryption effects and can effectively counteract the common attacks. Moreover, the scheme holds promising application prospects due to its authentication capability.

中文翻译：

---

具有计算重影成像的光学图像加密和认证方案

考虑到大多数场景下身份认证的必要性，提出一种基于计算重影成像（CGI）的双通道（数字通道和光通道）图像加密与认证算法。整个加密过程由4D混沌系统和Logistic映射获得的伪随机序列控制。数字通道加密过程中，明文图像的混淆和扩散分别通过随机置乱和DNA计算来实现。为了减轻CGI加密系统中密钥传输和存储的负担，利用Logistic映射得到的伪随机数构造散斑测量矩阵。此外，还进行了非线性加密操作，以弥补CGI加密的线性漏洞。在光学加密通道中，首先使用CGI方法对认证图像进行加密，然后对桶检测器记录的强度值进行二次加密，产生中间密文。最终的加密图像是通过合并来自两个加密通道的中间密文生成的。结果表明，所提出的基于CGI的光学图像加密和认证方案提供了令人满意的加密和解密效果，并且能够有效对抗常见的攻击。此外，该方案由于其认证能力而具有广阔的应用前景。