This month we have an article from a left field journal; "Journal of Data and Information Quality (JDIQ)". The latest issue has an article entitled "Compensated Signature Embedding for Multimedia Content Authentication" which proposes a new watermarking-based framework to conduct multimedia content authentication. The aim of this proposed authentication framework is to ensure the originality and quality of online multimedia contents. It is an interesting article and I wonder how such technological developments would affect the accessibility of such multimedia content.

Compensated Signature Embedding for Multimedia Content Authentication

One of the main goals of digital content authentication and preservation techniques is to guarantee the originality and quality of the information. In this article, robust watermarking is used to embed content-based fragile signatures in multimedia signals to achieve efficient authentication without requiring any third-party reference or side information. To overcome the signature alteration caused by the embedding perturbation and other possible encoding operations, a closed-form compensation technique is proposed for ensuring signature consistency by employing a Lagrangian-based approach. A minimum distortion criterion is used to ensure signal quality. The effectiveness of the proposed approach is investigated with simulations of examples of image authentication in which signatures are designed to reveal tamper localization. Results using quantitative performance criteria show successful authentication over a range of robustness in embedding watermarks using both QIM-DM and spread-spectrum techniques. A comparison with two iterative compensation schemes is also presented.

Full Paper: http://doi.acm.org/10.1145/1659225.1659230
Full Proceedings: Journal of Data and Information Quality (JDIQ), Volume 1, Issue 3, December 2009.

Labels: data quality, information quality, multimedia

Have you ever used myacm.org? If you are an ACM member, myacm.org includes a number of personalised services, for example e-mail forwarding, filtering, e-alerts, etc. It also lists the most popular articles in the ACM Digital library and also the most popular searches. According to their data, the most popular article is A feature-based algorithm for detecting and classifying scene breaks and was downloaded 2402 times. This paper describes an approach to detecting and classifying scene breaks. I think the work in this paper could be useful to people working on video accessibility.

A Feature-Based Algorithm for Detecting and Classifying Scene Breaks

We describe a new approach to the detection and classification of production effects in video sequences. Our method can detect and classify a variety of effects, including cuts, fades, dissolves, wipes and captions, even in sequences involving significant motion. We detect the appearance of intensity edges that are distant from edges in the previous frame. A global motion computation is used to handle camera or object motion. The algorithm we propose withstands JPEG and MPEG artifacts, even at high compression rates. Experimental evidence demonstrates that our method can detect and classify production effects that are difficult to detect with previous approaches.

Full Paper: http://doi.acm.org/10.1145/217279.215266
Full Proceedings: Proceedings of the third ACM international conference on Multimedia, San Francisco, California, 1995.

Labels: multimedia, myacm, video accessibility