No abstract

Currently, the field of smart-* (home, city, health, tourism, etc.) is naturally heterogeneous and multimedia oriented. In such a domain, there is an increasing usage of heterogeneous mobile devices, as well as captors transmitting data (IoT). They are highly connected and can be used for many different services, such as to monitor, to analyze and to display information to users. In this context, data management and adaptation in real time are becoming a challenging task. More precisely, at one time, it is necessary to handle in a dynamic, intelligent and transparent framework various data provided by multiple devices with several modalities. This paper presents a Kali-Smart platform, which is an autonomic semantic-based context-aware platform. It is based on semantic web technologies and a middleware providing autonomy and reasoning facilities. Moreover, Kali-Smart is generic and, as a consequence, offers to users a flexible infrastructure where they can easily control various interaction modalities of their own situations. An experimental study has been made to evaluate the performance and feasibility of the proposed platform.

Image watermarking is a potentially effective and powerful solution for multimedia security. It is a technique that plays a fundamental role in protecting copyright, proofing ownership, and helps to authenticate sensitive data. Watermarking can cause a decrease in image quality, especially when it comes to medical images. This could provoke false diagnoses, which can lead to serious consequences on the patient's health. In fact, it becomes necessary to think for new more imperceptible, secure watermarking techniques. To provide such techniques, we proposed a novel blind watermarking scheme for medical images, based on Schur decomposition and chaotic sequence (CS). An efficient chaotic method is applied on the watermark and the cover image to obtain encrypted images divided into subblocks. A decomposition based on Schur is utilized to embed the encrypted watermark bits in the cyphered cover image blocks. The same CS is applied to extract the original watermark. We compared the proposed technique with other watermarking schemes. The experimentation results demonstrate that the suggested method achieving good image fidelity and acceptable robustness.

No abstract