Satoru Tezuka

PC-based real-time watermark embedding system with standard video interface

An improved system is described for embedding watermarks into video frames in real time using software on an ordinary personal computer. It uses standard video I/O and is separate from the encoding process, so it can be incorporated into various encoding and distributing systems, which makes it well suited for distributing live content. Real-time processing is achieved by making a watermark-pattern generation process common to every frame a pre-process and shortening the watermark embedding time by reusing the watermark pattern output from this pre-process. It is also achieved by storing the watermarked video frames into the video memory on the video I/O board, thereby eliminating the need for storing them in buffers on hard disk. Testing of a prototype system demonstrated the validity of this approach.

Use of Motion Estimation to Improve Video Watermarking for MPEG Encoders

Maintaining picture quality is an essential requirement for video watermarking, and this paper proposes a new method that uses motion estimation (block-matching techniques) to preferentially allocate watermarks to picture contents with motion based on our criterion for estimating watermark imperceptibility that takes motion into account. Experimental evaluations have shown that the new method significantly reduces degradation in picture quality and that, for the same picture quality after MPEG encoding and decoding, 30% more watermarks can be embedded than with the previous method. This new method would be most effective when implemented in MPEG encoders, which already have the necessary motion estimation functions.

Real-time video watermark embedding system using software on personal computer

Wide use of broadband networks and high-efficiency personal computers facilitates distribution of live video content through the Internet. Digital watermarking for video helps protect the copyright of the video content, making real-time watermark embedding an essential requirement for content distribution. This paper describes a real-time embedding system suitable for live content distribution that uses software on an ordinary personal computer. Real-time processing is achieved by combining all the processes related to watermark embedding, and by shortening the watermark embedding time by reusing the estimated watermark imperceptibility among neighboring frames. A prototype system demonstrated the validity of the proposed system.

PERCEPTUALLY ADAPTIVE VIDEO WATERMARKING USING MOTION ESTIMATION

Video watermarks must maintain picture quality and withstand image processing, and resolving the tradeoff between these conflicting requirements has been one of the central problems in research on video watermarking. This paper shows that watermarks are less perceptible where picture contents are moving and proposes a new criterion for measuring watermark imperceptibility from motion vectors and deformation quantities. It also proposes a watermarking method that uses this criterion to allocate watermarks to picture areas adaptively. Experimental evaluation showed that the proposed method reduces degradation in picture quality and that, for the same picture quality after MPEG encoding and decoding, between 30% and 40% more watermarks can be embedded, improving detection reliability after MPEG processing by an average of 9.5%. The proposed method would be most effective when used with MPEG encoders because they already have the necessary motion estimation functions.

Print traceability systems framework using digital watermarks for binary images

Information systems still have security gaps at their analog output interfaces printing confidential information. Digital watermarking for binary image can be used for tracking a paper even if it has been photocopied. We clarify the technical requirement for print traceability systems and propose the basic concept of the system framework. Potential print traceability systems are categorized according to the identifiers that are embedded in the contents and that specify who, what, which, where, when, why or how. Our prototype system demonstrates that these traceability applications are ready for practical use.

Information Embedding Method for Home Printing of Certifications

In this paper, we describe our dot-pattern code, which can embed information onto the paper. Its main features are 1) no need to reserve a special room for the code, 2) obscure and unnoticeable, 3) comparatively large information capacity, 4) the absolute coordinates in the paper is measurable. We describe how to embed and detect the information onto / from the paper, and we estimate it by examination. From the examination, we found that our dot-pattern code is applicable to various kinds of papers, printers including ink-jet and scanners. We also found that even when the density of the black dot is greater than 6 %, over 8,000 bits could be safely embedded with error rate less than 10-6 . We also found that the embedded information could be detected correctly after the paper had been damaged.

Integrity Verification System For Video Content By Using Digital Watermarking

An improved system is described for verifying video content integrity using digital watermarking. Current verification systems using the digital signature are unable to distinguish between attacks and regular modifications such as resizing and MPEG encoding and are thus unsuitable countermeasures against actual threats to content kept in storage services. The proposed verification system distinguishes attacks against video content from regular modifications by extracting timecodes and header hash values embedded in the content itself and comparing them with the actual ones, making it well suited for content storage services. Evaluation showed that the system is more effective than a current one using the digital signature scheme and that it can be used by a variety of applications using stored video content

Seamless object authentication in different security policy domains

In the trading of intangible goods, there co-exist, from the security policy point of view, several different domains, such as CORBA and Java. In such environments, mobile objects such as Applets, can move freely between domains, which contributes greatly to the dynamic evolution of the distributed computer system. However, there arises a new security problem: there is no way for the mobile objects to move appropriately and seamlessly between the different security policy domains, according to the required rights. This cannot be prevented by the present security functions that are provided in such an environment and in each of the mobile objects. We therefore propose Seamless Object Authentication (SOA), which is: (1) a function which clearly defines the required rights to the mobile object, and (2) a function for the domains to check the mobile objects when they move between domains. We believe that this will lead to the dynamic evolution of the distributed computer system between different security policy domains and provide the basis of a Security Centric System.

Improved Video Verification Method Using Digital Watermarking

A method is described that verifies video content integrity by checking the continuity of embedded timecodes used as digital watermarks. Conventional verification methods using digital signatures and fragile watermarking are unable to distinguish between attacks and regular modifications due and thus are unable to protect against threats to content. The proposed verification method distinguishes attacks against video content from regular modifications by extracting timecodes embedded in consecutive frames of the content and then checking their continuity. A prototype implementation showed that the method is more effective than conventional ones and that it can be used by a variety of applications using video content.

A study on user authentication infrastructure for next generation telematics

To receive a telematics service, it must first be authenticated, and each automobile (or car navigation system) has typically been linked to a particular vehicle owner who could authenticate such services. However, an increasing number of drivers use cars that are rented, leased, or obtained through a car-sharing program; in such cases, the present form of car authentication is insufficient. A better system would authenticate the individual who is using a car at a particular time, but is not necessarily the owner. In this paper, we propose a personal authentication infrastructure for next-generation telematics. Specifically, we propose personal authentication into that (1) between the person and the terminal, (2) between the terminal and the center.