Mohamed Elhoseny

Bio: Mohamed Elhoseny is an academic researcher from Mansoura University. The author has contributed to research in topics: Computer science & Encryption. The author has an hindex of 49, co-authored 240 publications receiving 7044 citations. Previous affiliations of Mohamed Elhoseny include Maharaja Agrasen Institute of Technology & Cairo University.

Improving Sustainable Vegetation Indices Processing on Low-Cost Architectures

Abstract: The development of embedded systems in sustainable precision agriculture has provided an important benefit in terms of processing time and accuracy of results, which has influenced the revolution in this field of research. This paper presents a study on vegetation monitoring algorithms based on Normalized Green-Red Difference Index (NGRDI) and Visible Atmospherically Resistant Index (VARI) in agricultural areas using embedded systems. These algorithms include processing and pre-processing to increase the accuracy of sustainability monitoring. The proposed algorithm was evaluated on a real database in the Souss Massa region in Morocco. The collection of data was based on unmanned aerial vehicles images hand data using four different agricultural products. The results in terms of processing time have been implemented on several architectures: Desktop, Odroid XU4, Jetson Nano, and Raspberry. However, this paper introduces a thorough study of the Hardware/Software Co-Design approach to choose the most suitable system for our proposed algorithm that responds to the different temporal and architectural constraints. The evaluation proved that we could process 311 frames/s in the case of low resolution, which gives real-time processing for agricultural field monitoring applications. The evaluation of the proposed algorithm on several architectures has shown that the low-cost XU4 card gives the best results in terms of processing time, power consumption, and computation flexibility.

Return Rate Prediction in Blockchain Financial Products Using Deep Learning

Abstract: Recently, bitcoin-based blockchain technologies have received significant interest among investors. They have concentrated on the prediction of return and risk rates of the financial product. So, an automated tool to predict the return rate of bitcoin is needed for financial products. The recently designed machine learning and deep learning models pave the way for the return rate prediction process. In this aspect, this study develops an intelligent return rate predictive approach using deep learning for blockchain financial products (RRP-DLBFP). The proposed RRP-DLBFP technique involves designing a long short-term memory (LSTM) model for the predictive analysis of return rate. In addition, Adam optimizer is applied to optimally adjust the LSTM model’s hyperparameters, consequently increasing the predictive performance. The learning rate of the LSTM model is adjusted using the oppositional glowworm swarm optimization (OGSO) algorithm. The design of the OGSO algorithm to optimize the LSTM hyperparameters for bitcoin return rate prediction shows the novelty of the work. To ensure the supreme performance of the RRP-DLBFP technique, the Ethereum (ETH) return rate is chosen as the target, and the simulation results are investigated in different measures. The simulation outcomes highlighted the supremacy of the RRP-DLBFP technique over the current state of art techniques in terms of diverse evaluation parameters. For the MSE, the proposed RRP-DLBFP has 0.0435 and 0.0655 compared to an average of 0.6139 and 0.723 for compared methods in training and testing, respectively.

An Optimal Lightweight Cryptographic Hash Function for Secure Image Transmission in Wireless Sensor Networks

Abstract: In the recent years, numerous security schemes have been proposed to secure the data and Digital Images (DI) over WSNs. Especially, encryption and decryption algorithms are structured and actualized to provide secrecy and security in WSN during the transmission of image-based information just as in storage. In this chapter, Lightweight Cryptography (LWC) based hash function is used for image security in WSN. The hash function keeps up different guidelines which contain a set of tenets with user details, IP address, public and private keys. The hash value of encryption was developed upon the optimal secret key and it was recognized by the Enhanced Cuckoo Search (ECS) optimization. In this ECS model, cuckoo birds choose the nests of various birds to leave its eggs i.e., optimal keys. Further impressive fitness function parameters such as Peak Signal to Noise Ratio (PSNR) were kept consistent in this research. The proposed system provided expanded security and adequately utilized the algorithm when compared with ordinary encryption and optimization strategies.

Urban Scene Recognition via Deep Network Integration

Abstract: Scene understanding remains a challenging task due to the complex and ambiguous nature of scene images in defiance of several networks pre-trained on large-scale benchmark datasets are available. In this paper, we proposed a multi-scale, discriminative integrating method to aggregate both context and multi-scale object information of urban scene images. Our model is formulated as the integration of both patches and networks. Our scene-centric network is built upon the network fine-tuned on scenery dataset; our object-centric is an aggregation of multi-scale networks pre-trained on object-centric networks. We show that the integration of networks leads to an improvement in performance and achieved a 6.63% boost on overall accuracy comparing with the best-performed base model.

Self-organizing single-hop wireless sensor network using a genetic algorithm: Longer lifetimes and maximal throughputs

Abstract: In wireless sensor networks (WNSs), the network structure, which specifies how sensor nodes work, is greatly affects the network lifetime. To build the optimum network structure in cluster-based WSN may differs from round to round depending on a set of sensor nodes factors, i.e, remaining energy, vulnerability index, and the distance from BS. Getting the intended optimum structure is non trivial process, which includes determining the appropriate number of clusters, electing a cluster head (CH) for each cluster, and assigning each sensor node to a clusters. Recently, several studies propose CH selection protocols with predefined number of clusters. The present work introduces a new Genetic Algorithm (GA) based protocol that aims to get the optimum network structure for single-hop cluster model in WSN. This structure may differ after each round. The results show that the proposed GA-based method leads to more network lifetime and throughput. Also, it is more efficient in the context of a dynamic environment.

Computer vision : a modern approach = 计算机视觉 : 一种现代的方法

Abstract: From the Publisher: The accessible presentation of this book gives both a general view of the entire computer vision enterprise and also offers sufficient detail to be able to build useful applications. Users learn techniques that have proven to be useful by first-hand experience and a wide range of mathematical methods. A CD-ROM with every copy of the text contains source code for programming practice, color images, and illustrative movies. Comprehensive and up-to-date, this book includes essential topics that either reflect practical significance or are of theoretical importance. Topics are discussed in substantial and increasing depth. Application surveys describe numerous important application areas such as image based rendering and digital libraries. Many important algorithms broken down and illustrated in pseudo code. Appropriate for use by engineers as a comprehensive reference to the computer vision enterprise.

Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia the official statement of the American Thoracic Society and the Infectious Disease Society of America (特集 救急診療ガイドライン) -- (海外のガイドライン)

Abstract: Executive Summary Introduction Methodology Used to Prepare the Guideline Epidemiology Incidence Etiology Major Epidemiologic Points Pathogenesis Major Points for Pathogenesis Modifiable Risk Factors Intubation and Mechanical Ventilation Aspiration, Body Position, and Enteral Feeding Modulation of Colonization: Oral Antiseptics and Antibiotics Stress Bleeding Prophylaxis, Transfusion, and Glucose Control Major Points and Recommendations for Modifiable Risk Factors Diagnostic Testing Major Points and Recommendations for Diagnosis Diagnostic Strategies and Approaches Clinical Strategy Bacteriologic Strategy Recommended Diagnostic Strategy Major Points and Recommendations for Comparing Diagnostic Strategies Antibiotic Treatment of Hospital-acquired Pneumonia General Approach Initial Empiric Antibiotic Therapy Appropriate Antibiotic Selection and Adequate Dosing Local Instillation and Aerosolized Antibiotics Combination versus Monotherapy Duration of Therapy Major Points and Recommendations for Optimal Antibiotic Therapy Specific Antibiotic Regimens Antibiotic Heterogeneity and Antibiotic Cycling Response to Therapy Modification of Empiric Antibiotic Regimens Defining the Normal Pattern of Resolution Reasons for Deterioration or Nonresolution Evaluation of the Nonresponding Patient Major Points and Recommendations for Assessing Response to Therapy Suggested Performance Indicators