Catálogo de publicaciones - libros

The need for propagation mechanism in distributed environment is increasing in applications such as distributed databases, collaborative editing of Web documents, cooperative software developments etc. The propagation function required in these applications are often very different from the traditional approach. Hence, a flexible way of specifying and implementing propagation function on individual objects in distributed systems is required. We show how a propagation model with multi-class facilities can be constructed.

The motivation for the presented work is to make the propagation process as easy as possible. As a part of the data manipulation language, we have developed reusable operators for object propagation. The approach can be extended to object evolution in the context of mobile intelligent environment.

In this paper we deal with the problem of non-intuitive and low-efficient negotiations between agents in agent based system. We find recommendation techniques as a suitable method of making negotiation smarter and more efficient. We have introduced into the negotiation thread the following improvements: content-based recommendation and collaborative one. These two types of recommendation are discussed. On the base of a negotiation algorithm we have studied how introduced recommendation methods can improve the whole process of finding mutually acceptable agreements among agents.

This paper describes a method of using Petri net P-invariants in system diagnosis. The considered discrete event system is modelled by a live, bounded, and reversible place-transition net. The notions of D-partition of the set of places P of a given place-transition net N and net k-distinguishability are first introduced. Two different diagnosis test strategies are discussed and several examples are given.

The evolution for image manipulation tools and the wide distribution of images through the internet have led to the need for image copyright protection and forgery detection tools. In this paper a three models of a Robust-Fragile Dual watermarking system are proposed for the proof of ownership and image authentication. The proposed system has two main stages the embedding stage and the extraction stage. The embedding of the watermarks, which are the biometric signature for the image’s owner, is done in the DCT domain of the cover image. The proposed system with its three models is tested against attacks and it gives satisfactory results in the three models for robustness and fragility and data payload.

This paper presents a new approach for recovering blocks of missing data in the mobile radio environment. When block based compression algorithm such as JPEG are used as part of the wireless transmission process and such images are transmitted over fading channels, the noise effects will usually cause block missing or damaging. The proposed method is based on quantizing the difference of coefficients between two distinct blocks and carefully encoding and embedding them in the least significant bits of other blocks. The proposed scheme can recover high-quality JPEG images from the corresponding corrupted images up to a block loss rate of 20% without carrying any side information. Experimental results show the proposed method outperforms than the method of Shao et al.[1].

This paper investigates the transform domain-based adaptive watermark algorithm based on fuzzy inference. In addition, our approach has double-aim in first to enhance the robustness of the watermarking and in second to reduce the complexity of the fuzzy inference rule. Not only statistical characteristics (SC) but also human visual system (HVS) parameters, such as contrast sensitivity, texture degree and average value are involved to select the optimal coefficients region. Rule based fuzzy inference map is designed by utilizing a possible relation between HVS and SC. The performance analysis of the proposed approach is evaluated by the experiments of imperceptibility and correctness of watermark. Experimental results showed that proposed scheme has a better PSNR and has stronger response than previous results of Cox et al.

Recently, According as digital technology with knowledge base society develops, illegal copy by user’s vulnerability of copyright authentication becomes frequent. Therefore, it need research about user’s copyright authentication method. In this paper, a watermark inserting adaptive method according to the sensitivity of human visual system (HVS) in DCT transform domain is proposed. Also, we proposed watermark techniques using adaptive voice information(VI). It effects that get in original image than method to use conventional image is less and shows robust watermark restoration ability form outside attack. Performance analysis is achieved about still image and restoration of watermark information using OFDM/QPSK still image transmission system in wireless channel environment. Analysis result, VI watermark performance that influence in original image is very small. And it could know that show high restoration performance.

The rapid development of Internet and multimedia technologies has made copyright protection and multimedia retrieval be the two most important issues in the digital world. To solve these problems simultaneously, this paper presents a multipurpose watermarking scheme for image retrieval. First, several important features are computed offline for each image in the database. Then, the copyright, annotation and feature watermarks are offline embedded into all images in the database. During the online retrieval, the query image features are compared with the exacted features from each image in the database to find the similar images. The experimental results based on a database with 1000 images in 10 classes demonstrate the effectiveness of the proposed scheme.

In this paper, a novel watermarking scheme in the compressed domain is proposed for video copyright protection. The robust watermark is embedded in motion vectors of the bitstream based on the relationships between one-pixel accuracy and half-pixel accuracy motion vectors at the encoder. Watermark can be blindly extracted by parsing the properties of motion vectors at the decoder. Simulation results show that the proposed algorithm can perform the real-time watermark embedding process without affecting visual quality of the video sequence and it is also standard-compliant. The proposed algorithm can be applied to any video compression standard with half-pixel accuracy motion vectors such as MPEG-2 and H.263.

Digital audio watermarking may suffer from the different attacks simultaneously in DA/AD conversions, which may be modeled to noise corruption, modification of signal energy, and phase change. In this paper, we conduct many experiments to show that the phase change of the audio signals in DA/AD conversions may be represented as the temporal scaling, which results in the desynchronization of the watermark. Furthermore, we analyze the performance of quantization-based audio watermarking against Gaussian noises and the modification of signal amplitude. We present the corresponding calculation expression of BER . Finally, it is concluded that quantization-based audio watermarking is very susceptible to DA/AD conversions.