A High-Quality Authenticatable Visual Secret Sharing Scheme Using SGX

Denghui Zhang,Zhaoquan Gu,Zhuojun Duan

doi:10.1155/2021/6660709

Denghui Zhang, Zhaoquan Gu + Show 1 more

Open Access

https://doi.org/10.1155/2021/6660709

Copy DOI

Abstract

Visual cryptography scheme (VCS) is a secret-sharing scheme which encrypts images as shares and can decrypt shares without digital devices. Although a participant can reveal the secret image by merely stacking a sufficient number of shares, the visual quality of recovered images is reduced, and malicious adversaries can cheat participants by giving faked shares. The paper presents a novel VCS called T-VCS (trusted VCS) which consists of two main components: a high-quality VCS and an enhanced verification scheme of shares based on the emerging Intel Software Guard eXtensions (SGX). While providing high-quality recovery, T-VCS keeps the size of the shares the same as the original secret image. We use SGX to act as a trusted third party (TTP) to verify the validity of the shares in an attested enclave without degrading the image quality. The experimental results show that T-VCS can achieve a balance among contrast, share size, and verification efficiency.

Highlights

With the development of the Internet of things (IoT), wearable and mobile devices are forming more and more social and big data networks
The main contributions of this paper are as follows: (1) We propose a novel Visual cryptography scheme (VCS) that can both recover secret images with high quality and eliminate pixel expansion
(3) We elaborately evaluate the performance of applying T-VCS to encode grayscale and color images and verifying sheets in TEE

Summary

Introduction

With the development of the Internet of things (IoT), wearable and mobile devices are forming more and more social and big data networks. We propose a novel visual secret sharing scheme called T-VCS to address all the drawbacks listed above This method retains the advantage of traditional visual cryptography without any cryptography computation. (1) We propose a novel VCS that can both recover secret images with high quality and eliminate pixel expansion (2) We develop a self-attesting framework to verify the validity of shares based on TEE. This method neither requires the user to maintain additional verification data nor reduces image quality (3) We elaborately evaluate the performance of applying T-VCS to encode grayscale and color images and verifying sheets in TEE.

Related Work

Evaluation

Conclusions