Mohammad Al-Rousan

Spatio-Temporal Feature-Extraction Techniques for Isolated Gesture Recognition in Arabic Sign Language

This paper presents various spatio-temporal feature-extraction techniques with applications to online and offline recognitions of isolated Arabic Sign Language gestures. The temporal features of a video-based gesture are extracted through forward, backward, and bidirectional predictions. The prediction errors are thresholded and accumulated into one image that represents the motion of the sequence. The motion representation is then followed by spatial-domain feature extractions. As such, the temporal dependencies are eliminated and the whole video sequence is represented by a few coefficients. The linear separability of the extracted features is assessed, and its suitability for both parametric and nonparametric classification techniques is elaborated upon. The proposed feature-extraction scheme was complemented by simple classification techniques, namely, K nearest neighbor (KNN) and Bayesian, i.e., likelihood ratio, classifiers. Experimental results showed classification performance ranging from 97% to 100% recognition rates. To validate our proposed technique, we have conducted a series of experiments using the classical way of classifying data with temporal dependencies, namely, hidden Markov models (HMMs). Experimental results revealed that the proposed feature-extraction scheme combined with simple KNN or Bayesian classification yields comparable results to the classical HMM-based scheme. Moreover, since the proposed scheme compresses the motion information of an image sequence into a single image, it allows for using simple classification techniques where the temporal dimension is eliminated. This is actually advantageous for both computational and storage requirements of the classifier

Video-based signer-independent Arabic sign language recognition using hidden Markov models

Sign language in Arab World has been recently recognized and documented. There have been no serious attempts to develop a recognition system that can be used as a communication means between hearing-impaired and other people. This paper introduces the first automatic Arabic sign language (ArSL) recognition system based on hidden Markov models (HMMs). A large set of samples has been used to recognize 30 isolated words from the Standard Arabic sign language. The system operates in different modes including offline, online, signer-dependent, and signer-independent modes. Experimental results on using real ArSL data collected from deaf people demonstrate that the proposed system has high recognition rate for all modes. For signer-dependent case, the system obtains a word recognition rate of 98.13%, 96.74%, and 93.8%, on the training data in offline mode, on the test data in offline mode, and on the test data in online mode respectively. On the other hand, for signer-independent case the system obtains a word recognition rate of 94.2% and 90.6% for offline and online modes respectively. The system does not rely on the use of data gloves or other means as input devices, and it allows the deaf signers to perform gestures freely and naturally.

Recognition of Arabic sign language alphabet using polynomial classifiers

Building an accurate automatic sign language recognition system is of great importance in facilitating efficient communication with deaf people. In this paper, we propose the use of polynomial classifiers as a classification engine for the recognition of Arabic sign language (ArSL) alphabet. Polynomial classifiers have several advantages over other classifiers in that they do not require iterative training, and that they are highly computationally scalable with the number of classes. Based on polynomial classifiers, we have built an ArSL system and measured its performance using real ArSL data collected from deaf people. We show that the proposed system provides superior recognition results when compared with previously published results using ANFIS-based classification on the same dataset and feature extraction methodology. The comparison is shown in terms of the number of misclassified test patterns. The reduction in the rate of misclassified patterns was very significant. In particular, we have achieved a 36% reduction of misclassifications on the training data and 57% on the test data.

Implementation of experimental communication protocol for health monitoring of patients

Modern wireless communication technology has provided new tools for collecting data from remotely distributed sensors. Global System for Mobile Communications (GSM) services like General Packet Radio Service (GPRS) and Short Message Service (SMS) have proven to be legitimate and cost effective methods for collecting occasional data from remote locations. A communication protocol that facilitates remote data collection using SMS has been developed to collect data from large number of monitoring stations. In this study, the developed protocol is implemented and tested to monitor medical condition of large number of patients. The paper gives implementation details and the results of implementation.

Embedded system-based mobile patient monitoring device

This paper describes the design of a mobile medical device that can be used to monitor the human temperature and blood pressure (BP) using a stand-alone single chip microcontroller. The device hardware architecture consists of temperature and pressure sensors, signal conditioning circuits (SCC), single chip microcontroller, LCD display and GSM modem. An embedded software algorithm acquires temperature and pressure, processes, transmits, displays and stores it in the built-in EPROM of the microcontroller. A preset trigger level for the temperature and/or the BP is stored in the EEPROM of microcontroller. Once the desired programmed trigger level of any of the signal is reached, the microcontroller downloads the current value of the temperature and BP to the GSM modem. Then, GSM automatically dials presorted mobile number/s and transmits both parameters as a normal mobile message to a physician, nurse and/or emergency personal. A short database containing these parameters are collected and stored in a lookup table. This database can be used to track the patient temperature and BP history if needed.

Energy consumption considerations in dynamic wireless sensor networks with nodes and base stations mobility

Energy conservation is a critical issue in the design of sensor networks since the sensor nodes are battery-powered. This paper proposes a wireless sensor network with mobile sensor nodes and base stations to prolong the network lifetime. The movement of base stations inside the covered region is done according to the evolution of current events. The proposed scheme is based on minimising the average consumed energy for every active sensor when data is transmitted. Performance evaluation of the proposed wireless dynamic sensor network is done with a comparison of the network lifetime with the case of static sensor nodes and the base stations.

AMCC: Ad-hoc based Mobile Cloud Computing Modeling☆

Abstract In this paper a new ad-hoc model for mobile based cloud computing is proposed based on the cloudlet approach. The model is using the Destination-Sequenced Distance-Vector (DSDV) for routing protocol and Random Way Point (RWP) for mobility mechanism. The model aim at reducing the end-to-end packet delay, better system scalability and mobility management. The model is flexible with various workload sizes that are offloaded to cloudlets and for different nodes speed. It also, exploit the mobile devices ability in utilizing its context awareness such as its locations. The model is predicted to achieve a lower hand-off delay than using an enterprise cloud unless offloading small workload size.

A living biological nano robot as self-navigator sensor for diseases

In this paper, we experimentally investigated the navigation system of the nonpathogenic strain of E. coli (AW405) and developed a simulator for the locomotion performance of these swimming nanorobots. The swimming behavior of these cells is sensitive to the chemical gradients in their medium. Tissue and disease cells might produce chemical signals in their surroundings. These chemicals have the potential to affect the locomotion behavior of the bacterial cells. Therefore, bacterial cells can be considered as self-navigator nanorobots that are able to discriminate between disease cells such as cancer. Our current experimental and theoretical work is considered as a platform to this novel idea of early detection of problematic diseases.

An innovative m‐utility management and billing system

Purpose – Aims to provide a new wireless design for utility management and billing systems using GSM networks that can be used by large utility sites.Design/methodology/approach – Traditional energy and water meters are replaced or enhanced to produce digital readings. A hardware interface is designed and then connected via the GSM network to a server at the utility headquarters. The new system is compared with traditional utility systems.Findings – A prototype for the proposed systems was implemented. It was tested using a real GSM network in the United Arab Emirates. The performance of the systems was acceptable, with high accuracy results when compared with the existing systems.Research limitations/implications – Applying the system in large buildings with multiple subscribers is not a straightforward task; the system may need some modifications.Practical implications – This is a very useful system for utility companies which are interested in better management and billing systems. The system can be im...

Real-Time Communications for Wireless Sensor Networks: A Two-Tiered Architecture

While many approaches have been proposed to deal with energy/latency trade-offs, they are likely to be insufficient for the applications where reduced delay guarantee is the main concern. In this article, we investigated the potential application of a decentralized two-tiered network architecture, in large-scale wireless sensor networks, where an upper layer Wireless Local Area Network (WLAN), offering more powerful capabilities, serves as a backbone to an adaptively-clustered Low-Energy Adaptive Clustering Hierarch (LEACH)-based wireless sensor network. The WLAN layer will be involved in the communication between the sensor network and the control station, mitigating the impact of the limited capacities of the sensor nodes. With this two-tiered architecture we target to provide more reliable data delivery with reduced delay bounds, and lower energy consumption in the underlying sensor network, thereby increasing its lifetime. Simulation results show that the two-tiered network architecture achieved a relatively long lifetime, while preserving remarkably low latencies, compared to a single-tiered LEACH and a super-clustered LEACH-based network architectures.