Catálogo de publicaciones - libros

In this paper we propose a video in video hiding technique for embedding and transmission of sign language video signal through another video communication channel. Our proposed system does not need modification of the main video coding and transmission system, and it can be implemented by addition of a pre- and post-processing block. We encode the sign language video and embed the encoded bit-stream in the main host video. We develop a low complexity and robust encoding scheme for sign language video by applying three-dimensional wavelet decomposition on the sign language video, and encoding the important subbands using multiple description scalar quantizers. We also apply three-dimensional wavelet decomposition on the host video and embed the encoded bit-stream of the sign language video in the host signal subbands. We select blocks of the host video with high texture content and embed the information in the middle frequency subbands to have highest robustness and with low visible distortion. At the receiver the system reconstructs the sign language video by recovering the embedded data in subbands. We examine the system performance by embedding video sequences of sign language messages into host video sequences. At the receiver we evaluate the percentage of correct sign language messages recovery. We further evaluate the system performance when the host video undergone compression and addition of noise during transmission.

Piracy is a major issue in the content business, and although digital technologies raise the efficiency of new content production, pirates also accelerate their piracy. Commercial cameras have become sufficiently good to re-shoot images displayed on screens, and TV monitors are becoming larger with better picture quality; this combination will enable content to be pirated in the home in the near future. Re-shooting with a commercial camera is the ultimate piracy method because there is no engineering way to stop it. We have developed a new watermark method that remains even after images displayed on a monitor have been re-shot. This paper describes the concept and experimental results.

A semi-fragile watermarking algorithm for digital image content authentication is proposed in this paper. The watermark is generated from the Zernike moments of the low frequency subband in IWT (integer wavelet transform) domain of the original image. This sort of image-dependent watermark can resist forgery attack and improve the watermark security. Watermark is embedded in the IWT domain of an image by exploiting the features of human visual system (HVS). It makes the high-fidelity and high-quality watermarked image. Moreover, we can distinguish mild distortion from severe distortion according to the difference of Zernike moments before and after attack. Experimental results show that the proposed algorithm is fragile to malicious tamper and can locate the tampered area accurately while is robust to content-preserved manipulation, such as JPEG lossy compression.

A new watermarking scheme which provides the ability of sharing secret with multi-users is proposed. It splits the original watermark into two shares and embeds one share into the cover image to increase the security. A polarization procedure is performed to establish a polarity stream from the cover image. The second share and the polarity stream are used to generate a master key and several normal keys. In our system, only the super user can reveal the genuine watermark directly. Other users possess the normal keys can obtain shadow watermarks merely. By combining the shadow watermarks together, the real watermark can be recovered.

We have proposed a novel audio watermarking technology, which embeds a set of bitstream data by changing two-channel stereo locations of lower frequency components in an embedding target audio signal. This method features nearly lossless embedding, robustness against lossy data compression or analogue conversion, and enables contactless asynchronous detection of embedded watermarks through speaker and microphone devices without the original audio signals. In this paper, we propose an extended monaural audio watermark embedding method, which supports both monaural and stereo audio signals, and enables watermark detection by a single monaural microphone such as cell phone voice receiver devices. We describe an abstract of our proposed watermark embedding and extracting algorithm, and experimental results of watermark extraction precision.

In this paper, a novel application of SVG (Scalable Vector Graphics) is proposed in watermarking. Vector graphic is used to detect geometric transform and control geometric distortion. Experimental results show that the proposed method is efficient to solve geometric attack to watermarking.

Invertibility attack is a hostile measure to breach watermarking systems. In this paper, a novel watermarking protocol using a one-way hash function and a check of random watermarks is proposed in order to combat invertibility attacks. The described technique can be used in conjunction with any watermarking algorithm, no matter it is kept secret or made public, without resorting to a third party jury as required by some previous approaches. By introducing a set of reference sequences, segmentation of the digital information and iterative computation of watermarks, the protocol is further enhanced so that it can resist more sophisticated types of attack based on forging an illegitimate detector.

A hybrid watermark that consists of a robust part and a fragile part can be used to serve multiple purposes. The robust part can protect copyright information, the fragile part can detect tampering, and their combination enables identification of attacks encountered. This paper analyses an overlap and a non-overlap implementation of the robust and fragile parts in a hybrid system. The difference between the two implementation methods lies in the robust and fragile watermarks embedding positions. Embedding capacity, computational costs, watermark robustness, and tamper detection localization of the two implementations are analyzed. In addition, optimization issues of block size in the hybrid system are discussed.

In this paper, we propose a new secret sharing scheme (SSS) [1] for audio signals, called as “Binary audio secret sharing (BASS).” SSS is an encryption method and produces n shared data from an original data to hide useful information. Applying SSS to audio communications on the Internet can help to make it more robust against theft and the tapping of information. Thus, we focused on the 1-bit audio format and applied SSS to 1-bit audio signals to realize audio secret sharing. Moreover, we propose a method to make each shared data heard as its intended sound.

A new watermarking technique called associative watermarking is proposed in this paper. The watermark in our new scheme will be embedded by applying the concept of association rules borrowed from the field of data mining. We implant the watermark’s rules into the original image by changing a very small portion of the image’s pixels during the rules alignment process. As a result, the robustness can be enhanced by emphasizing the specific rules mined from critical areas in the original image. Instead of transforming the watermark into its bit pattern, we just mine the association rules from the watermark and embed them into a colorful image. Our experimental results reveal that the PSNR value of the image after watermark implanting exceeds 40, and more than 90% of the watermarked images can be accurately recognized after similarity checking.