Speckle‐Based Optical Cryptosystem and its Application for Human Face Recognition via Deep Learning

Abstract

Face recognition has become ubiquitous for authentication or security purposes. Meanwhile, there are increasing concerns about the privacy of face images, which are sensitive biometric data and should be protected. Software-based cryptosystems are widely adopted to encrypt face images, but the security level is limited by insufficient digital secret key length or computing power. Hardware-based optical cryptosystems can generate enormously longer secret keys and enable encryption at light speed, but most reported optical methods, such as double random phase encryption, are less compatible with other systems due to system complexity. In this study, a plain yet highly efficient speckle-based optical cryptosystem is proposed and implemented. A scattering ground glass is exploited to generate physical secret keys of 17.2 gigabit length and encrypt face images via seemingly random optical speckles at light speed. Face images can then be decrypted from random speckles by a well-trained decryption neural network, such that face recognition can be realized with up to 98% accuracy. Furthermore, attack analyses are carried out to show the cryptosystem's security. Due to its high security, fast speed, and low cost, the speckle-based optical cryptosystem is suitable for practical applications and can inspire other high-security cryptosystems.

High-security learning-based optical encryption assisted by disordered metasurface

“…As discussed in our previous work 21 , speckle-based cryptosystem benefits from the complexity of the physical secret key demonstrating high-level security. Nevertheless, if the ciphertext (i.e., speckles) is accidentally obtained by the hackers, it is expected that the system still has the ability to protect itself.…”

“…Another lens (L2) put in front of the camera is used to adjust the grain size of the recorded speckles. Sine the decryption is not a trivial inverse of the scattering process like other works 16 , 20 , 21 (more detailed discussion will be given in “Discussion”), a DNN named DMNet is specifically designed to match the physical process, with details provided in Supplementary Note 3 .…”

“…4a, b : mismatch of the QR code fails to decrypt the image). Such a scheme substantially differentiates the learning-based decryption process in our work from the peer works whose ciphertext (speckle pattern) contains only one component originating from the phase object, in which high-level security is hard to ensured 16 , 20 , 21 . In the decryption process, only when the speckle, the security key, and the polarization are all correct and matched, the plaintext can be deciphered correctly.…”

Wavefront shaping improves the transparency of the scattering media: a review

“…It is worth mentioning that although dealing with scattering in biomedicine is the current focus, the applications of WFS are not limited to this. 118 – 121 As mentioned in the introduction, in recent years, holographic display 122 – 124 and optical communication 125 – 127 have also been supported by the rapid development of WFS. However, at present, they are still in the experimental verification stage.…”

Wavefront shaping: A versatile tool to conquer multiple scattering in multidisciplinary fields

“… 204 , 205 Moreover, advances in optimization algorithms and recent applications of deep learning are expected to play a crucial role in realizing faster and more efficient WFS. 88 , 206 , 207 Looking back and looking forward, we believe that WFS holds a bright future that will open up new avenues for noninvasive or minimally invasive optical interactions and arbitrary control inside deep tissues. The high degree of freedom in multiple scattering and WFS will also provide unprecedented opportunities to develop novel optical devices based on a single scattering medium (generic or customized) that can outperform traditional optical components.…”