Mohamed Sedky

Classification of smart video surveillance systems for commercial applications

Video surveillance has a large market as the number of installed cameras around us can show. There are immediate commercial needs for smart video surveillance systems that can make use of the existing camera network (e.g. CCTV) for more intelligent security systems and to contribute in more applications (beside or) rather than security applications. This work introduces a new classification for smart video surveillance systems depending on their commercial applications. This paper highlights different links between the research and the commercial applications. The work reported here has both research and commercial motivations. Our goals are first to define a generic model of smart video surveillance systems that can meet requirements of strong commercial applications. Our second goal is to categorize different smart video surveillance applications and to relate capabilities of computer vision algorithms to the requirement of commercial application.

OpenSHS: Open Smart Home Simulator

This paper develops a new hybrid, open-source, cross-platform 3D smart home simulator, OpenSHS, for dataset generation. OpenSHS offers an opportunity for researchers in the field of the Internet of Things (IoT) and machine learning to test and evaluate their models. Following a hybrid approach, OpenSHS combines advantages from both interactive and model-based approaches. This approach reduces the time and efforts required to generate simulated smart home datasets. We have designed a replication algorithm for extending and expanding a dataset. A small sample dataset produced, by OpenSHS, can be extended without affecting the logical order of the events. The replication provides a solution for generating large representative smart home datasets. We have built an extensible library of smart devices that facilitates the simulation of current and future smart home environments. Our tool divides the dataset generation process into three distinct phases: first design: the researcher designs the initial virtual environment by building the home, importing smart devices and creating contexts; second, simulation: the participant simulates his/her context-specific events; and third, aggregation: the researcher applies the replication algorithm to generate the final dataset. We conducted a study to assess the ease of use of our tool on the System Usability Scale (SUS).

A self-aware paradigm for autonomous architectural systems within the Internet of Things

Abstract This research explores a newfangled approach for the notion that architecture is no longer concerned with the organization of space and matter, but rather a system of systems with self-organizational behavior and progressively complex programmatic interventions. Such system represents an Internet of Things (IoT) paradigm that has the capacity to self-organize a pre-defined architecture acclimating dynamically to the demands of the environment through a process of self-awareness and re-configurability. This paper is an attempt to develop spaces that bring computation into the physical world by introducing artificial intelligence into building systems so as to communicate, exchange information, and allow right responses and decisions within a sustainable manner. Environments with embedded computational systems and adaptive reconfiguration behavior will be precisely studied. With an intensive focus on smart, self reconfiguring architecture that has embraced kinetic motion as an approach for environmental adaptation and responsiveness. This study addresses the networked organization of sensory systems that incorporate computational platforms in relation to user’s desires, where architecture turns into an interactive intermediary between human and computation. Correspondingly, the authors are particularly aiming to propose three conceptual frameworks for delivering a smart system at the architectural scale which is capable of reconfiguring and interacting constantly in real time, precisely focusing on an initial implementation for the first framework utilizing an experimental customized prototype, while the other two frameworks would be subsequently studied through future work. Thus the IoT will foster a rapid development in intelligent systems which would directly introduce an advanced technological leap forward in accommodating a new way of demonstrating human-computer interaction within a novel architectural design approach. The figure introduces three diverse approaches for reconfiguring an architectural space –for a smart campus- through the Internet of Things as an autonomous paradigm for anticipating and receiving parameters of particular inputs within a relevant context of students’ interaction with university programs.

Video analysis for Yellow Box Junction violation: Requirements, challenges and solutions

In recent years, a series of efforts have been undertaken in order to bring more intelligence to the public transportations of smart cities to solve essential problems like traffic management for public safety, smart traffic monitoring and congestion management. In this paper, we propose solutions for the Yellow Box Junction violation in order to reduce traffic congestion on roads by using video analytic techniques and a model for traffic monitoring using a single and dual camera (Fixed and PTZ Camera) as master-slave control. This paper highlights the major challenges related to computer vision which are encountered in the design of traffic management systems. Further enhancement in traffic management can be done by designing efficient feature extraction, tracking and recognition algorithms that can help in improving the classification results of the proposed system.

Physics-based and Retina-inspired Technique for Image Enhancement

This paper develops a novel image/video enhancement technique that integrates a physics-based image formation model, the dichromatic model, with a retina-inspired computational model, multiscale model of adaptation. In particular, physics-based features (e.g. Power Spectral Distribution of the dominant illuminant in the scene and the Surface Spectral Reflectance of the objects contained in the image are estimated and are used as inputs to the multiscale model for adaptation. The results show that our technique can adapt itself to scene variations such as a change in illumination, scene structure, camera position and shadowing and gives superior performance over the original model.