Region-based transform-domain video scrambling

Abstract

In this paper, we address the problem of scrambling regions of interest in a video sequence. We target applications such as video surveillance preserving privacy, anonymous video communications, or TV news safeguarding the anonymity of a source. We propose an efficient solution based on transform-domain scrambling. More specifically, the sign of selected transform coefficients is flipped during encoding. We address the two cases of Motion JPEG 2000 and MPEG-4. Simulation results show that it can be successfully applied to conceal information in regions of interest in the scene while providing with a good level of security. Furthermore, the scrambling is flexible and allows adjusting the amount of distortion introduced. Finally, this is achieved with a small impact on coding performance and negligible computational complexity increase.

A survey on JPEG2000 encryption

“…We observe that the SSIM index is capable of measuring the similarity even for very low quality images in contrast to the PSNR and the ESS (see Figs. 8,9). These results also relativize the claim of data confidentiality for the technology example of Annex B.10 [32, p. 85] (in this example only 1% of JPEG2000 data is encrypted).…”

“…The signs of the wavelet coefficients are scrambled in a number of contributions [9][10][11][12][13], mainly with the goal to preserve privacy. In [11], as well as in [10], flipping the signs of selected coefficients is proposed for "privacy enabling technology for video surveillance".…”

Privacy enabling technology for video surveillance

“…Motion JPEG 2000 is an extension of JPEG 2000 for the coding of video sequences which intra-frame encodes each frame using DWT-based JPEG 2000 [6]. However, the approach is extensible to other motion compensated transform-coding techniques such as Motion JPEG, MPEG-4 or AVC as shown in [7].…”

“…We consider more specifically the case of Motion JPEG 2000, an extension of JPEG 2000 for the coding of video sequences, which consists of intra-frame coding of each frame using Discrete Wavelet Transform (DWT) JPEG 2000 [6]. However, the method is extensible and can be used with most existing video coding standards, such as Motion JPEG, MPEG-4 or Advanced Video Coding (AVC) as shown in [7]. Video analysis is used to identify regions corresponding to people and assumed to contain privacy-sensitive information.…”

Transparent encryption for HEVC using bit-stream-based selective coefficient sign encryption

“…Although numerous approaches for DCT-based video coding standards like MPEG-2 Video, MPEG-4 Part 2 and H.264 have been proposed [4,5,6,7,8,9,10], the latter flip all AC coefficient signs to encrypt the content, aiming at full encryption. In contrast, our approach selectively flips AC coefficient signs of the luminance channel, reducing the quality slightly, but noticeably, allowing for transparent encryption.…”