In this paper, we propose a secure watermarking scheme based on spread transform dither modulation (STDM) method for digital cinema. The embedding is performed in the JPEG2000 decoding pipeline after the de-quantization and prior to the inverse discrete wavelet transform (IDWT). We exploit the wavelet properties related to the human visual system (HVS) in order to have a trade-off between fidelity and robustness, while preserving security. We design a pixel-wise masking vector that modulates the spreading vector in such a way that preserves its security. Our results show that the proposed method is robust against traditional image processing attacks. The proposed scheme can also survive the camecording attack, a pre-processing step is done in the detection for this end.
Darazi, R., Callau, P., & Macq, B. (2009). Secure and HVS-adaptive exhibition spread transform dither modulation watermarking for digital cinema. 2009 First IEEE International Workshop on Information Forensics and Security (WIFS 2009), p. 1-5. https://doi.org/10.1109/WIFS.2009.5386493