Digital Watermarks for Videos Based on a Locality-Sensitive Hashing Algorithm

Sensitive information in images is can be leaked during attacks, resulting in the malicious acquisition of personal information. To improve the robustness of attacking defence for video images, a digital watermarking algorithm based on locality-sensitive hashing (LSH) is designed in this paper. Video signals are decomposed using a one-dimensional wavelet transform. According to the Yeung Mintzer (Y-M) algorithm, a marker watermark W1 is embedded in the low-frequency subband to identify any image tampering. A data string of hash function values and the exclusive OR (XOR) result of identification watermark W2 are embedded into the HH high-frequency subband, which was used to identify and counter pseudo-authentication attacks such as collage and Vector Quantization (VQ). The singular value decomposition (SVD) algorithm is used to decompose the hash-mapped watermark and adaptively adjust embedding strength of the watermark. The position-sensitive hash algorithm which is proposed in this work has good invisibility for embedding digital watermarks into images, with an average accuracy of approximately 97% for feature matching of digital images. The PSNR value of the image embedded with the watermark is approximately 49 dB. At the 50th minute of the experiment, the regulatory factor value of the research method was 0.3. Under different attack modes, the correlation coefficient between the watermark extracted by this method and the original watermark image is greater than 0.85. Due to the low compression quality of JPGE, the correlation coefficient between the watermark and the initial watermark is greater than 0.6, and its error rate is less than 0.10 bits, showing the effectiveness of the proposed methodologies.

Automated summary

This paper proposes a digital watermarking algorithm based on locality-sensitive hashing to improve the robustness of attacking defence for video images. The experimental results demonstrate that the proposed method has good invisibility and high accuracy in identifying and countering image tampering and pseudo-authentication attacks.