On Screen-Shooting Leak Tracking Algorithm

Leaking data displayed on the screen through smartphones or photographic apparatus has become the main means of modern leaks. Most of the existing methods study how to prevent data from being photographed from the perspective of defense. To achieve effective tracking of data leakers, this paper proposes a digital-watermarking-based screen-shooting leak tracking algorithm. Specifically, we first propose a watermark generation algorithm, which can generate a watermark that uniquely determines the user's identity and ensures the non-repudiation of the watermark. Furthermore, to ensure the good transparency of the watermark, an encoder that can embed the watermark into the carrier image is designed. Finally, to ensure the robustness of the watermark, we design a set of noise layers to simulate the distortion during the screen-shooting process and use it to train the decoder to ensure the accuracy of tracking. A large number of experiments show that compared with other mainstream algorithms, the algorithm proposed in this paper has a high tracking accuracy in various screen-shooting processes, and can effectively track the leaker.