Screen-Shooting Resilient Watermarking Scheme via Learned Invariant Keypoints and QT.

Li Li,Shanqing Zhang,Mengtao Shi,Rui Bai,Chin-Chen Chang

doi:10.3390/s21196554

Abstract

This paper proposes a screen-shooting resilient watermarking scheme via learned invariant keypoints and QT; that is, if the watermarked image is displayed on the screen and captured by a camera, the watermark can be still extracted from the photo. A screen-shooting resilient watermarking algorithm should meet the following two basic requirements: robust keypoints and a robust watermark algorithm. In our case, we embedded watermarks by combining the feature region filtering model to SuperPoint (FRFS) neural networks, quaternion discrete Fourier transform (QDFT), and tensor decomposition (TD). First we applied FRFS to locate the embedding feature regions which are decided by the keypoints that survive screen-shooting. Second, we structured watermark embedding regions centered at keypoints. Third, the watermarks were embedded by the QDFT and TD (QT) algorithm, which is robust for capturing process attacks. In a partial shooting scenario, the watermark is repeatedly embedded into different regions in an image to enhance robustness. Finally, we extracted the watermarks from at least one region at the extraction stage. The experimental results showed that the proposed scheme is very robust for camera shooting (including partial shooting) different shooting scenarios, and special attacks. Moreover, the efficient mechanism of screen-shooting resilient watermarking could have propietary protection and leak tracing applications.

Highlights

Taking photographs has become highly efficient and convenient as shown by the widespread use of smart phone, pinhole, augmented reality (AR) and virtual reality (VR)cameras as well as mini digital video recorders (DVRs)
We propose a screen-shooting watermarking scheme via learned invariant keypoints, which combines feature region filtering model to SuperPoint (FRFS), quaternion discrete Fourier transform (QDFT), and tensor decomposition (TD) (FRFSQT)
We introduce the double-transformation watermarking algorithm based on TD and QDFT in the frequency domain

Summary

Introduction

Taking photographs has become highly efficient and convenient as shown by the widespread use of smart phone, pinhole, augmented reality (AR) and virtual reality (VR). Cameras as well as mini digital video recorders (DVRs). This new efficiency could pose a threat to information security field. Capturing computer screen photos and videos is an important means of stealing internal confidential information, which is difficult to prohibit and leaves no trace. A robust watermarking scheme that can extract information from screen-shot photos should be designed to protect confidential information. We can embed identifying information, such as a screen or user identifer number, in the host image, and through the extracted message, we can provide a reliable way to authenticate an images and protect copyright.

Methods

Results

Conclusion