JPEG‐based scalable privacy protection and image data utility preservation

Abstract

Here, the authors propose a scalable scrambling algorithm operating in the discrete cosine transform (DCT) domain within the JPEG codec. The goal is to ensure that people are no more identifiable while keeping their actions still understandable regardless of the image size. For each 8 × 8 block, the authors encrypt the DCT coefficients to protect data information, and shift them towards the high frequencies to make the DC position available. Whereas encrypted coefficients appear as noise in the protected image, the DC position is dedicated to restitute some of the original information (e.g. the average colour associated with one or a group of blocks). The proposed approach automatically sets the value of each DC according to the region of interest size in order to keep the level of privacy protection strong enough. Comparing to existing methods, the proposed privacy protection framework provides flexibility concerning the appearance of the protected version which makes it stronger for protecting the privacy even during potential attacks. Moreover, the method does not cause excessive perturbation for the recognition of the actions and slightly decreases the efficiency of the JPEG standard.

Reversible Data Hiding in Jpeg Images for Privacy Protection

“…In the privacy preserving algorithms for JPEG images based on the DCT scrambling (e.g., [8,22,23]), the protected region will be irregularly deformed with block effects. To improve visual quality of the protected regions as well as achieve smaller file size, the first 6 DCT coefficients of the luminance component (including the DC and AC ones) in zigzag order are modified, which is also enough to protect the content of selected regions according to the experimental results.…”

“…According to the work in [3], the block number ranges from 25 to 1600 in a protected region while most are within 25 to 400. Besides, the experiments in [22] show that there are 26 nonzero AC coefficients in a block on average. Given a protected region containing 25 blocks with 26 nonzero coefficients per block, the number of combinations to recover the flipped AC coefficients is 2 25 * 26 = 2 650 , showing that the proposed algorithm is resistant to brute force attack.…”

Content and Privacy Protection in JPEG Images by Reversible Visual Transformation

“…Nevertheless, F5 steganography [19] or another JPEG steganography algorithm (e.g., [20]) is applied in the embedding operation, but the original JPEG images cannot be recovered completely. Similarly, a scalable scrambling algorithm operating in the DCT domain is proposed in [13] within the JPEG codec. The goal is to ensure that people are no more identifiable while keeping their actions still understandable regardless of the image size.…”

“…It can be seen that the block number ranged from 25 to 1600, while most of them were concentrated between 25 and 400. Besides, the experiments in [13] show that there are 26 nonzero AC coefficients in a block on average. Suppose that the attacker knows the location of the protected regions and the algorithm adopted for visual transformation.…”

“…In the literature, a number of methods (e.g., [2][3][4][5][6][7][8][9][10][11][12][13]) have been proposed for privacy protection in JPEG images, including selective encryption (e.g., [2]), image blurring, image pixelation, image masking, image morphing (e.g., [3]) and image warping (e.g., [4]). To keep the sensitive content from being exposed, these methods may change image content permanently and even degrade image quality.…”