Video Watermarking Technique using Visual Sensibility and Motion Vector

An adaptive compressed domain watermarking method is proposed in the paper. In this technique, multiple binary images derived from a single watermark image are first embedded in a video sequence. The spatial spread spectrum watermark is directly incorporated in the compressed bit streams by modifying the wavelet coefficients. Experimental evidence demonstrates that the developed approach is efficient and also robust against spatial attacks such as scaling and frame averaging, and temporal attacks like frame dropping and shifting.

To achieve robustness and imperceptibility, an adaptive compressed domain blind video watermarking method based on Discrete Wavelet Transform (DWT) is proposed in this research. In this technique, multiple binary images derived from a single watermark image are first embedded in a video sequence. The spatial spread spectrum watermark is directly incorporated in the compressed bit streams by modifying the four sets of discrete wavelet coefficients. Comprehensive simulation experiments demonstrate that the developed approach is efficient and also robust against spatial attacks such as scaling and frame averaging, noise attacks such as Gaussian and salt pepper noise, and temporal attacks like frame dropping and shifting. Moreover, the proposed approach can also withstand against rotation attacks of arbitrary angle.

The fast progress in digital techniques and internet utilization brought new problems viz. copyright protection and authentication of the multimedia data namely image, audio and video. The watermarking technique is widely used as a potential solution to address these issues by incorporating additional information called "watermark" into the data of multimedia. An optimal hybrid watermarking method is introduced by exploiting the properties of nonsubsampled contourlet transform (NSCT) and singular value decomposition (SVD) for uncompressed videos in this paper. The proposed video watermarking algorithm is an additive and non-blind, which necessitates the presence of every frame of original video for watermark extraction / detection. In this approach, the original video frame and the watermark are decomposed simultaneously by nonsubsampled contourlet transform. The singular values of each frame of original video are altered with that of singular values of corresponding watermark with appropriate scaling factor. Here, different watermarks are incorporated in different scenes of original video to enhance the robustness for attacks. Experimental results show that the combination of NSCT-SVD based hybrid watermarking algorithms exhibit a better imperceptibility and robustness to various attacks, namely, rotation, cropping, color to gray scale conversion, salt and pepper noise, and frame dropping. The performance of this method is examined with the existing watermarking methods, and observed a better performance in quality of video frame and robustness to various attacks.

the paper proposes a tool, VisiMark1_0, as assistance for evaluating the robustness of video watermarking algorithms as evaluation of a video watermarking algorithm for robustness with available tools is a tedious task. It is our belief that for researchers of robust video watermarking, a tool needs to be designed that will assist in the evaluation procedure irrespective of the design algorithm. This tool provides a test bed of various attacks. The input to this tool will be a watermarked video whereas the outputs will be attacked videos, evaluated parameters PSNR, MSE, MSAD and DELTA, graphical comparisons of the attacked and watermarked videos with all parameters needed by researchers, and the attacks report. Provision for comparison of any two videos is an additional facility provided in the tool. The attacks implemented in VisiMark1_0 are categorized mainly in three. Firstly, Video attacks: Frame dropping, Frame averaging, Frame swapping, Changing the sequence of the scenes, Changing Frame rate, Fade and dissolve, Contrast stretching, Motion blurring, Chroma sampling, Inter frame averaging are some of the novel offerings in video frame attacks category. Secondly, Geometrical attacks: Apart from the traditional Rotation, Scaling and Cropping attacks for images, VisiMark1_0 contributed towards Sharpening, Shearing, Flipping, Up/down sampling and Dithering attacks for a video and signal processing attacks like Conventional Noising, Denoising and Filtering attacks for images are incorporated for video along with Pixel removal attack as a novel contribution. VisiMark1_0 is an endeavor to design a tool for evaluating a raw video (an .avi file currently) incorporating various attacks having a prospect for numerous video formats in near future.

Digital video is one of the popular multimedia data exchanged in the internet. Due to their perfectly replicable nature many illegal copies of the original video can be made. Methods are needed to protect copyrights of the owner and prevent illegal copying. A video can also undergo several intentional attacks like frame dropping, averaging, cropping and median filtering and unintentional attacks like addition of noise and compression which can compromise the owner information thereby denying authentication. In this paper, a robust Discrete Wavelet Transform (DWT)-based blind digital video watermarking scheme with scrambled watermarks based on scene changes has been proposed for authentication of digital video, which embeds different parts of a single watermark into different scenes of a video. The video watermarking algorithm is robust against the attacks of frame dropping, averaging and compression, which are considered as some of the common types of attacks applied particularly on the video and due to the use of DWT it can also withstand geometrical attacks making the watermark perceptually invisible. Furthermore, it allows blind retrieval of embedded watermark which does not need the original video. The proposed algorithm has been compared with an existing DWT based watermarking scheme and is found to exhibit better robustness.

In this paper, a robust algorithm of digital image watermarking based on discrete wavelet transform is introduced It uses blind watermarking technique. Digital image watermarking is one such technology that has been developed to protect digital images from illegal manipulations. In particular, digital image watermarking algorithms which are based on the discrete wavelet transform have been widely recognized to be more prevalent than others. This is due to the wavelets' excellent spatial localization, frequency spread, and multi-resolution characteristics, which are similar to the theoretical models of the human visual system.

Based on the existing image and video watermarking algorithms, this paper proposes a new digital video watermarking algorithm based on DCT domain. The basic idea of this algorithm is to use the difference between two DCT coefficients in the DCT domain to embed watermark information in consecutive frames of the original video. The simulation experiment results show that the algorithm has better invisibility.

Infrared images often involve trade secrets. In order to protect the information security of infrared images, using a robust watermarking algorithm based on infrared images is a fitting solution. Recently, many effective visible and infrared image fusion (VIF) algorithms have been proposed in medicine, biology, and geology. Robust watermarking algorithms can resist mild conventional attacks, and with the progress of science and technology, algorithmic robustness against novel attacks such as screen shooting and print shooting has also become a research hotspot. However, VIF‐based image watermarking algorithms are still scarce. It is important to investigate a robust watermarking algorithm that can resist VIF attacks. Herein, an autoencoder against infrared fusion attacks is proposed based on various subnetworks: CRDnet (Convolutional Residual Dense Network). CRDnet includes encoders and decoders based on residual and dense structures, a fusion network robust to 12 VIF algorithms, and predictors for predicting watermarked infrared images. The encoder and decoder also incorporate preprocessing steps, attention mechanisms, and activation functions suitable for infrared images. The experimental results demonstrate that the bit error rate of CRDnet is reduced by at least ≈4% compared to common autoencoders. The peak signal‐to‐noise ratio of watermarked images is also almost always greater than 38 dB.

A new video watermarking algorithm based on 1D DFT and Radon transform

A digital image-watermarking algorithm based on fractional Fourier transform (FRFT) domain is presented by utilizing chaotic CDMA technique in this paper. As a popular and typical transmission technique, CDMA has many advantages such as privacy, anti-jamming and low power spectral density, which can provide robustness against image distortions and malicious attempts to remove or tamper with the watermark. A super-hybrid chaotic map, with good auto-correlation and cross-correlation characteristics, is adopted to produce many quasi-orthogonal codes (QOC) that can replace the periodic PN-code used in traditional CDAM system. The watermarking data is divided into a lot of segments that correspond to different chaotic QOC respectively and are modulated into the CDMA watermarking data embedded into low-frequency amplitude coefficients of FRFT domain of the cover image. During watermark detection, each chaotic QOC extracts its corresponding watermarking segment by calculating correlation coefficients between chaotic QOC and watermarked data of the detected image. The CDMA technique not only can enhance the robustness of watermark but also can compress the data of the modulated watermark. Experimental results show that the watermarking algorithm has good performances in three aspects: better imperceptibility, anti-attack robustness and security.

With the development and mature of the Internet and wireless communication techniques the copy and distribution of the digital production becomes easier and faster. The copyright protection is an urgent problem to be resolved. The watermark is a digital code unremovably, robustly, and imperceptibly embedded in the host data and typically contains information about origin, status, and/or destination of the data. Most of present watermarking algorithms embed only one watermark, however one watermark is not sufficient under some circumstances. One of the reasons of privacy is that we cannot trace the responsibility of the pirates. It would come true by means of embedding the different exclusive watermarks belong to the issuer. As the different watermarks are needed at the different time, the multiple watermarks algorithm is required. The reports about multiple watermarks scheme are rather few. Cox et al. extend the single watermark algorithm to embed multiple orthogonal watermarks. The disadvantage is that the original image is needed at the watermark detector, and the watermark capacity is small. Stankovic et al. proposed a scheme utilizing the two-dimensional Radon–Wigner distribution. The lack is that the watermark capacity is small and we cannot judge the validity of the extracted watermark directly. Tao et al. present a multiple watermark algorithm in the DWT (Discrete Wavelet Transform) domain. The binary image is used for the watermark, and is embedded into all frequencies of DWT. The shortage is that the algorithm is not blind.

A real-time video watermarking scheme against geometric distortions is proposed for DCT-encoded compressed video data. The full DCT coefficients have proven to be invariant to scaling and local geometric attacks. Therefore, watermarks are embedded into watermark minimal sequences (WMS) by modulating the low-frequency full DCT coefficients. To meet the requirement of real-time performance, a fast intertransformation is employed to construct full DCT directly from block DCTs. The modified differences of full DCT coefficients are inversely transformed to the differences of block DCT coefficients that are subsequently added to the original block DCTs to generate the WMS. To resist various video format conversions, the watermark detection is performed in spatial domain. This does not affect the real-time performance because the watermark detection need not re-encode the frames into compressed data, and compressed video can be easily decoded fast by employing some codec or decoding chips. Furthermore, the proposed scheme can resist rotation attacks by employing a rotation compensation strategy. Experimental results demonstrate that the proposed scheme is real-time, transparent, and robust to signal distortions, geometric distortions including rotation, scaling, aspect ratio change, linear geometric transforms, cropping and combination attacks, frame dropping and swapping, file format conversion, as well as camcorder recording.

본 논문에서는 동영상 콘텐츠의 저작권 보호를 위한 워터마크의 삽입 공간을 충분히 확보할 수 있도록 3차원 웨이블릿 변환과 대역확산 상관도의 위치 코딩을 이용하는 비디오 워터마킹 알고리즘을 제안하였다. 워터마크 정보 삽입을 위해 위치 코딩을 이용하고 정보의 삽입과 추출을 <TEX>$1{\times}1$</TEX>화소 단위가 아닌 <TEX>$4{\times}4$</TEX> 샘플 단위로 함으로써, 비트 에러나 회전에 대해 기존 방법보다 우수한 강인성을 얻었다. 실험결과 강도 11에서 압축하지 않은 상태의 PSNR은 동영상의 히스토그램이 협소한 경우에 최소 31.07dB로 비가시성을 만족하였으며, 6개 동영상에 대해 각각 3가지 코덱을 적용한 결과 모두 BER=0%의 결과를 얻음으로써 워터마크의 강인성을 확인할 수 있었다. In this paper, a new video watermarking algorithm using 3D wavelet transform and position encoding of spread spectrum correlation is proposed as a method for ensuring enough payload in order to protect copyright for video contents. This algorithm provides better robustness than other watermarking methods in bit error and rotation because it uses the position encoding and <TEX>$4{\times}4$</TEX> sample space instead of <TEX>$1{\times}1$</TEX> pixel to express a digit. The evaluation result over 6 video clips shows that the PSNR is over 31dB at strength 11 when histogram is narrow and it satisfies the transparency. Also, it shows the robustness of the proposed watermark algorithm that the BER is 0% to three codecs.

Research in digital watermarking has evolved rapidly in the current decade. This evolution brought various different methods and algorithms for watermarking digital images and videos. Introduced methods in the field varies from weak to robust according to how tolerant the method is implemented to keep the existence of the watermark in the presence of attacks. Rotation attacks applied to the watermarked media is one of the serious attacks which many, if not most, algorithms cannot survive. In this paper, a new automatic rotation recovery algorithm is proposed. This algorithm can be plugged to any image or video watermarking algorithm extraction component. The main job for this method is to detect the geometrical distortion happens to the watermarked image/images sequence; recover the distorted scene to its original state in a blind and automatic way and then send it to be used by the extraction procedure. The work is limited to have a recovery process to zero padded rotations for now, cropped images after rotation is left as future work. The proposed algorithm is tested on top of extraction component. Both recovery accuracy and the extracted watermarks accuracy showed high performance level.

Digital watermarking is an emerging technique to protect data security and intellectual property right. Identification or verification of watermarking patterns can be achieved by detecting watermarks in received signals. However, one of the biggest challenges in watermarking detection is that the strengths of the watermark signals will change after being distorted by an attacker in a watermarking channel. Meanwhile, the embedding strengths may be adapted to the original signals, which are unknown at the receiver end. Further, the original signals are often highly non-Gaussian. Although some work has been done on optimum detection of watermarks, the uncertainty of watermark signal strengths and real statistical behavior of multimedia contents have not been taken into account simultaneously. Much more study is needed to enhance the performance of watermarking systems. Since multiplicative watermarks are robust and well suited for copyright protection, this paper presents our investigation on robust optimum detection of multiplicative watermarks. For sub-band transformed domains such as the discrete cosine transform (DCT), discrete wavelet transform (DWT), and pyramid transform, a class of generalized correlators is constructed based on the generalized Gaussian distributions. Thresholding methods to achieve a given false alarm rate, and the performance analyses are provided. The square-root detector is designed and demonstrated to have near performance for a large set of natural images and can be employed as a universally optimal detector or decoder for images and video. The locally most powerful detection method is then extended to DFT domain multiplicative watermarking, with the magnitudes of coefficients modeled by the Weibull distributions. Another class of detectors is built based on this statistical modeling. The robust optimum detection of multiplicative watermarks can be applied to copyright notification, enforcement, and broadcast monitoring. We have applied the robust optimum watermarking detection to combined audio and video watermarking.