Alejandro Velez Zea

Three-dimensional joint transform correlator cryptosystem

We introduce for the first time, to the best of our knowledge, a three-dimensional experimental joint transform correlator (JTC) cryptosystem allowing the encryption of information for any 3D object, and as an additional novel feature, a second 3D object plays the role of the encoding key. While the JTC architecture is normally used to process 2D data, in this work, we envisage a technique that allows the use of this architecture to protect 3D data. The encrypted object information is contained in the joint power spectrum. We register the key object as a digital off-axis Fourier hologram. The encryption procedure is done optically, while the decryption is carried out by means of a virtual optical system, allowing for flexible implementation of the proposal. We present experimental results to demonstrate the validity and feasibility of the method.

Customized data container for improved performance in optical cryptosystems

Coherent optical encryption procedures introduce speckle noise to the output, limiting many practical applications. Until now the only method available to avoid this noise is to codify the information to be processed into a container that is encrypted instead of the original data. Although the decrypted container presents the noise due to the optical processing, their features remain recognizable enough to allow decoding, bringing the original information free of any kind of degradation. The first adopted containers were the quick response (QR) codes. However, the limitations of optical encryption procedures and the features of QR codes imply that in practice only simple codes containing small amounts of data can be processed without large experimental requirements. In order to overcome this problem, we introduce the first tailor made container to be processed in optical cryptosystems, ensuring larger noise tolerance and the ability to process more information with less experimental requirements. We present both simulations and experimental results to demonstrate the advantages of our proposal.

Optimized random phase only holograms

We propose a simple and efficient technique capable of generating Fourier phase only holograms with a reconstruction quality similar to the results obtained with the Gerchberg-Saxton (G-S) algorithm. Our proposal is to use the traditional G-S algorithm to optimize a random phase pattern for the resolution, pixel size, and target size of the general optical system without any specific amplitude data. This produces an optimized random phase (ORAP), which is used for fast generation of phase only holograms of arbitrary amplitude targets. This ORAP needs to be generated only once for a given optical system, avoiding the need for costly iterative algorithms for each new target. We show numerical and experimental results confirming the validity of the proposal.

Optimized random phase encryption

We propose for the first time, to the best of our knowledge, the use of optimized random phases (ORAPs) in a double random phase encryption scheme (DRPE). In DRPE schemes the convolution between two random phase functions encrypts the information to be secured. However, in actual encryption applications, this convolution of random phases also results in unwanted effects like speckle noise. In this Letter we show that under certain conditions this noise can be drastically reduced. These conditions can be easily achieved by using ORAPs. These ORAPs, besides containing information about the parameters of the optical system and maintaining all the security properties of a random phase function, ensure that the encrypted data is a phase-only function. This leads to a great increase in system performance, with decryption quality similar to the reconstruction of a phase-only hologram generated with the Gerchberg-Saxton algorithm. We show both numerical and experimental results confirming the validity of our proposal.