Intra-frame distortion drift is a big problem of data hiding in H.264/AVC video streams. Based on thorough investigation this problem, novel readable data-hiding algorithm, which can embed into the quantized discrete cosine transform (DCT) coefficients I frames without bringing any intra-frame H.264/advanced coding (AVC) host, presented paper. We exploit several paired-coefficients 4 <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML"...

Privacy becomes one of the major concerns video surveillance systems, especially in cloud-based systems. protection videos aims to protect privacy information without hampering normal tasks. Region-of-interest (ROI) is more practical compared with whole encryption approaches. However, common drawback virtually all current ROI methods that original compressed recorded camera permanently distorted by process, due quantization re-encoding process. Thus, integrity captured destroyed. This...

In practice, charts are widely stored as bitmap images. Although easily consumed by humans, they not convenient for other uses. For example, changing the chart style or type a data value in image practically requires creating completely new chart, which is often time-consuming and error-prone process. To assist these tasks, many approaches have been proposed to automatically extract information from images with computer vision machine learning techniques. achieved promising preliminary...

A novel data-hiding scheme with limited intra-frame distortion drift is proposed to hide information in H.264/AVC digital video. The secret message embedded into the quantized DCT (discrete cosine transform) coefficients of 4×4 residual luma blocks I frames. By analysis relationship between and pixel values used prediction, we choose suitable for embedding compensate by adjusting another coefficient. experimental results show that our can effectively control get higher visual quality....

Privacy becomes one of the major concerns cloud-based multimedia applications such as cloud video surveillance. protection surveillance videos aims to protect privacy information without hampering normal processing tasks cloud. Region Protection only protects region while keeping non-privacy visually intact facilitate in However, full reversibility, i.e. complete recovery original which is critical digital investigation and law enforcement has not been properly addressed protection. In this...

There is greater possibility that reversible video data hiding scheme causes more distortions than image hiding. In H.264/AVC streams, intra-frame distortion drift a big problem in This paper proposes novel readable which can avert the drift. We embed into quantized discrete cosine transform (DCT) coefficients of I frames meet directions prediction, and prediction are utilized to It proved analytically shown experimentally proposed algorithm achieve high embedding capacity low visual...

Fingerprint authentication has been extensively employed in contemporary identity verification systems owing to its rapidity and cost-effectiveness. Due widespread use, fingerprint leakage may cause sensitive information theft, enormous economic personnel losses, even a potential compromise of national security. As that can coincidentally match specific proportion the overall population, MasterPrint rings alarm bells for security authentication. In this paper, we propose new side-channel...

Privacy-preserving information utilization has attracted a lot of attention in the big data era where owners and processers may be separate entities. Cloud video surveillance is typical example that requires privacy-preserving processing. Video can leverage cloud for its storage, processing power, accessibility from anywhere with any devices. In such system, cameras simply stream to cloud, which then stores data, performs motion detection other analysis, alerts users if needed. presents...

In the cloud era, a large amount of data is uploaded to and processed by public clouds. The risk privacy leakage has become major concern for users. Cloud-based video surveillance requires motion detection, which may reveal people in video. Privacy-preserving allows detection while protecting privacy. existing scheme [ 25 ], designed detect on encrypted H.264-compressed videos, does not work well more advanced compression schemes such as HEVC. this article, we propose first method...

In many cases, a network may be dispersedly recorded by different participants and each participant is one part of the original network, no willing to share its data due commercial competition. Therefore, forms an independent private they form " multiple networks" regarding network. Existing methods only use structure itself predict missing links, leading underutilized information deteriorated prediction accuracy. One natural question arises: how integrate networks formulating security...

Abstract User alignment across online social network platforms (OSNPs) is a growing concern with the rapid development of internet technology. In reality, users tend to register different accounts on multiple OSNPs, and are reluctant share structure user’s information due business interest privacy protection, which brings great obstacles cross-platform user alignment. view this, we propose homomorphic encryption-based (HE-SNA) algorithm from perspective leakage. Specifically, first consider...

A novel data-hiding scheme without intra-frame distortion drift in H.264/AVC video streams is proposed this paper. We exploit several coefficients of a 4 × DCT block an I frame to accumulate the embedding induced and use directions prediction avert drift. And then we recover original as much possible when hidden data extracted out. The experimental results show that our can effectively drift, get high visual quality large capacity.

Cloud-based video surveillance requires protecting privacy yet allowing cloud to perform motion detection and tracking. Prior art allows performing on encrypted videos but details of trajectories moving objects are exposed. In this paper, we propose a system that supports adjustable granularity tracking for fine-grained control conflicting requirements between protection accuracy Our encryption adds permutation layer conventional format-compliant selective ensure information can be recovered...

Deep neural networks (DNNs) are vulnerable to adversarial attacks. A great effort has been directed developing effective defenses against attacks and finding vulnerabilities of proposed defenses. recently defense called Trapdoor-enabled Detection (TeD) deliberately injects trapdoors into DNN models trap detect examples targeting categories protected by TeD. TeD can effectively existing state-of-the-art In this paper, we propose a novel black-box attack on TeD, Feature-Indistinguishable...

This paper presents an improved H.264/AVC data hiding method based on without Distortion Drift technique and BCH syndrome coding technique. The in this is that we exploit several paired-coefficients of a 4 × DCT block to accumulate the embedding induced distortion use directions intra-frame prediction avert drift. encode embedded by before data. experimental results show new robust algorithm can effectively drift, get higher visual quality more robust.

Deep learning-based image embedding methods are typically designed for natural images and may not work chart due to their homogeneous regions, which lack variations hide data both robustly imperceptibly. In this paper, we propose ChartStamp, the first method that is robust real-world printing displaying (printed on paper displayed screen, respectively, then captured with a camera) while maintaining good perceptual quality. ChartStamp hides 100, 1,000, or 10,000 raw bits into image, depending...