University of Wah Journal of Computer Science

QoS-Aware Efficient Tasks Scheduling in Heterogeneous Cloud Computing Environment

Junaid Hassan, Zeshan Iqbal — Fri, 29 Dec 2023 00:00:00 +0000

The management and prioritization of network traffic to ensure the efficient transmission of important data is achieved through Quality of Service (QoS) technologies and techniques. QoS facilitates allocating necessary resources and bandwidth to critical applications and services while de-prioritizing less important traffic. This is accomplished by classifying and marking packets. Task scheduling involves coordinating and managing the execution of tasks in a computer system or network, including allocating resources and determining the order in which tasks are executed. Task scheduling algorithms use priority, resource requirements, and task dependencies to determine the most efficient way to execute tasks. A heterogeneous cloud environment utilizes multiple cloud computing platforms from different vendors such as IaaS, PaaS, and SaaS to deliver services and optimize cost, performance, and scalability. The task scheduling problem in cloud computing involves effectively mapping workloads to virtual resources. The study introduces the genetic-based algorithm BGA to increase makespan and resource consumption, while SGA focuses on convergence speed. These strategies are compared with current meta-heuristic and heuristic techniques.

Keywords: Cloud Computing; Task Scheduling; Genetic Algorithm; Quality of Service; Bandwidth

Secure Medical Imaging Data using Cryptography with Classification

Saqqiya Waris, Syeda Aleena Naqvi — Fri, 29 Dec 2023 00:00:00 +0000

Medical imaging data is increasing day by day which requires improved applications that perform accurate diagnoses. Secure medical imaging data plays a critical role in current times. Still, today it is a complex task to maintain data privacy, so this study's main objective is to solve this problem. In this research, firstly, secure the medical imaging data using the cryptography Advance Encryption Standard (AES) algorithm. In this process, input images are encrypted and decrypted using public key cryptography and supplied as input to the pre-trained convolutional neural Network such as Alex-net. The model comprises 25 layers such as convolutional, batch-normalization, ReLU and max-pooling, etc. The classification between the tumor and healthy images has been performed using the SoftMax layer. The performance of the proposed model has been tested on the publicly available BRATS-2020 Challenging dataset. The proposed model achieved up to 99.71% accuracy and 97% F1- scores, which are far better as compared to the latest published research work in this domain.

Keywords: Image Classification; Preprocessing; Features extraction; Cryptography; MRI images

A Comprehensive Review of Plant Disease Detection Using Deep Learning

Usra naz, Mehak Mushtaq Malik — Fri, 29 Dec 2023 00:00:00 +0000

The research on maize diseases is described in this study of current literature. The most valuable findings are extracted from researchers' previous work and presented in a compiled form. The article discusses the problem of detecting plant diseases using deep learning techniques. Plant diseases can cause significant damage to crops, and early detection is essential for effective treatment. The study demonstrates how the model and approach may be designed from a new viewpoint, which can then lead to improved outcomes. Following this, datasets that are accessible to the public are described and identified, and then the proposed procedures and the findings are tested and verified using these datasets. Performance metrics along with their respective formulae are discussed to demonstrate how these measures might be used to evaluate the effectiveness of research activity.

Keywords: Deep Learning; Computer Vision; Machine learning; Corn; Plant Diseases

Advancements In Artificial Intelligence for Brain Tumor Detection: A Comprehensive Survey

Rameesha Zia — Fri, 29 Dec 2023 00:00:00 +0000

Brain tumors are a group of diseases characterized by abnormal growths of cells in the brain that can cause severe neurological symptoms. In recent years, the advent of artificial intelligence (AI) techniques has shown great promise in enhancing brain tumor detection. This survey research discusses methods and techniques used for AI-based brain tumor detection. Brain tumors pose significant health risks, necessitating accurate and timely detection for effective treatment. The study defines brain tumors and emphasizes the need for precise detection methods due to tumor variations. Biomarkers associated with brain tumors are investigated, highlighting their potential as diagnostic and prognostic indicators. The utilization of deep learning (DL) models, including Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and 3D CNNs, is examined, providing a comparative analysis of their strengths and limitations. The importance of datasets, such as TCIA, BRATS, and ISLES, is discussed in training and evaluating AI models for brain tumor detection. This survey aims to contribute to the understanding and progress of AI-based brain tumor detection along with the comparison of some deep learning models, providing insights for researchers and healthcare professionals working towards improving patient outcomes.

Keywords: Brain tumor; Artificial Intelligence; Tumor detection; Diagnosis; Image Preprocessing; Deep Learning; Neural Networks, Molecular biomarkers.

A Novel CNN-RNN Model for E-Cheating Detection Based on Video Surveillance

Aqeel Zaffar, Muhammad Jawad; Muzammil Shabbir — Fri, 29 Dec 2023 00:00:00 +0000

Nowadays, everything needs to be digitized, and scientific knowledge is constantly bringing comfort and change to everyday life. Autonomous systems have become a prominent technology in recent years, with a variety of applications in different fields. Our proposed system is designed to maintain academic integrity in exams. The system uses computer vision techniques to monitor the behavior of students during the exam and detect any suspicious activities, such as looking at someone else's paper or using unauthorized materials. The proposed E-Cheating Detection consists of four core steps: 1) Student/Person Detection and Tracking, 2) Detect suspicious activities, 3) Generating alerts, and 4) Mark attendance. Student detection from videos is performed by using YOLOv7 and DeepSort tracker is used to track the detected persons that are being detected by the YOLOv7 algorithm. To classify suspicious activities (such as the exchange of paper, and giving codes to one another, etc.) the system uses CNN-RNN architecture in which the inceptionV3 model is used for feature extraction. The system will generate real-time alerts for suspicious behavior by sending an email via SMTP to the exam administration /invigilator. The system marks the student's attendance by recognizing and matching student faces that are stored in the database. The performance of the system will be evaluated by conducting a series of experiments using simulated scenarios, and the results will demonstrate the effectiveness of the proposed system in detecting suspicious activities during physical exams. The proposed system has the potential to promote exam integrity and create a fair environment for all students, ultimately improving the education system's quality.

Keywords: Detect Suspicious Activities; CNN-RNN, YOLOv7; DeepSort; Face Recognition; Physical Exam; Smart Attendance