.png){kind=icon}
.png){kind=icon}
Return
Optical encryption using scattering and computational ghost imaging
.png){kind=icon}
PRE
AI
DOI：10.1016/j.optlastec.2025.113919.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In this paper, we propose a multi-image optical encryption scheme that combines deep learning-assisted optical scattering encryption and ghost encoding. Specifically, two secret images are encrypted by the optical scattering system into speckle patterns and then sparsified. A pre-trained network model is used to reconstruct images from the sparse speckles, which are then utilized to generate the sampling region for ghost imaging encryption. Subsequently, the bucket signal obtained from the ghost encoding of the third image and the sparse speckle are combined into ciphertext through numerical processing. During the decryption process, two sparse speckle patterns are initially extracted from the ciphertext, which are then input into the neural network to restore the two original images. With the aid of these recovered images and a binarization threshold key, a sampling region map is generated, which facilitates the reconstruction of the third image from the bucket signal extracted from the ciphertext. Notably, in this scheme, the two encryption methods are highly integrated rather than being mutually exclusive. This integration not only enhances the encryption capacity but also enhances security. The experimental results validate the success of this attempt to concurrently incorporate scattering imaging and ghost imaging for encryption purposes.
Journal
O
IF:
5
Papers:
1.9K
Citations:
3.5W
Organization
No organization information available