Title: Digital watermarking of compressed videos using larger dimension 2D error correcting codes for higher embedding capacity

Authors: Anjana Rodrigues; Archana Bhise

Addresses: Department of Electronics and Telecommunication, NMIMS Deemed to be University, Mumbai, India ' Department of Electronics and Telecommunication, NMIMS Deemed to be University, Mumbai, India

Abstract: This paper proposes a novel method of digital watermarking of MPEG videos using 2-dimensional error-correcting codes (2D ECC). The motion vectors of the video to be protected are used as the cover. Hence, this method is versatile and can be used to watermark videos of multiple formats such as MPEG-1, MPEG-2, MPEG-4 (AVC/H.264) and even the latest format HEVC/H.265. An error-correcting code of dimension 15 × 15 is constructed and used to embed the watermark into select motion vectors at the encoder and to retrieve the watermark at the decoder. The use of 2D ECC facilitates embedding of multiple bits of the watermark in various patterns inside the cover, thus improving imperceptibility. The results obtained on sample videos show a high embedding capacity of 8 bits/pixel and 245 bits/codeword, while still maintaining a PSNR greater than 40 dB, as compared to the existing 4 bits/pixel of other methods.

Keywords: 2D error correcting codes? digital watermarking? video watermarking? MPEG videos? copyright protection? motion vectors.

DOI: 10.1504/IJESDF.2021.118546

International Journal of Electronic Security and Digital Forensics, 2021 Vol.13 No.6, pp.652 - 670

Received: 08 Jun 2020
Accepted: 10 Oct 2020
Published online: 28 Oct 2021 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article