International Journal on Recent and Innovation Trends in Computing and Communication
Not a member yet
8613 research outputs found
Sort by
ECG Biometric for Human Authentication using Hybrid Method
Get PDFRecently there is more usage of deep learning in biometrics. Electrocardiogram (ECG) for person authentication is not the exception. However the performance of the deep learning networks purely relay on the datasets and trainings, In this work we propose a fusion of pretrained Convolutional Neural Networks (CNN) such as Googlenet with SVM for person authentication using there ECG as biometric. The one dimensional ECG signals are filtered and converted into a standard size with suitable format before it is used to train the networks. An evaluation of performances shows the good results with the pre-trained network that is Googlenet. The accuracy results reveal that the proposed fusion method outperforms with an average accuracy of 95.0%
Newton’s Law of Gravitational Force (NLGF) based Machine Learning Technique for Uneven Illuminated Face Detection
Get PDFA photo gallery is crucial for organizing your photos, presenting them in beautiful categories, and doing sophisticated memory searches. The photo gallery is portrayed in a vocabulary of nonlinear similarities to the prototype face image collection. One of the difficult research ideas for machine learning technologies is the maintenance of a photo gallery using facial recognition. Based on changes in the faces' appearance, faces are identified. This research proposes novel machine learning algorithms to recognize faces by characterizing the majority of discriminating local characteristics, which maximizes the dissimilarity between face photos of different persons and reduces the dissimilarity between features between face images of the same person. This method relies on Newton's third law of gravitational force to determine the relationship between pixels to extract the features of noisy accurately and efficiently, unevenly illuminated, and rotationally invariant face images
Performance Analysis of Reputation based Proof of Credibility Consensus Mechanism for Blockchain based Applications
Get PDFBlockchain is a decentralized transaction and data management technology first developed for the Bitcoin cryptocurrency. Blockchain technology is gaining popularity due to its core attributes which provides security, anonymity and data integrity without any involvement of third party. Consensus mechanism is a procedure by which all peers in the blockchain network agrees to a common agreement on the current state of the distributed ledger. It plays vital role in increasing efficiency of any blockchain environment. Though we have many consensus mechanisms working currently in different areas but they still lack in parameters like status of validators, latency, node failure etc. In Our proposed algorithm Proof of credibility, we have tried to incorporate all above factors in it. We have also implemented two or more factors of proposed algorithm and have evaluated and compared with existing consensus algorithm. In future research we aim to implement RPoC in any blockchain network and then we will evaluate it in terms of different evaluation parameters such as performance, security, scalability
Enhancing Retinal Scan Classification: A Comparative Study of Transfer Learning and Ensemble Techniques
Get PDFOphthalmic diseases are a significant health concern globally, causing visual impairment and blindness in millions of people, particularly in dispersed populations. Among these diseases, retinal fundus diseases are a leading cause of irreversible vision loss, and early diagnosis and treatment can prevent this outcome. Retinal fundus scans have become an indispensable tool for doctors to diagnose multiple ocular diseases simultaneously. In this paper, the results of a variety of deep learning models (DenseNet-201, ResNet125V2, XceptionNet, EfficientNet-B7, MobileNetV2, and EfficientNetV2M) and ensemble learning approaches are presented, which can accurately detect 20 common fundus diseases by analyzing retinal fundus scan images. The proposed model is able to achieve a remarkable accuracy of 96.98% for risk classification and 76.92% for multi-disease detection, demonstrating its potential for use in clinical settings. By utilizing the proposed model, doctors can provide swift and accurate diagnoses to patients, improving their chances of receiving timely treatment and preserving their vision
EETSQ: Energy Efficient Task Scheduling based on QoS Parameters in Cloud Computing Environment
Get PDFNow a day, energy consumption is the big challenge in heterogeneous cloud computing environment that needs to be considered. Cloud service provider also needs to satisfy customer’s Quality of Service (QoS) for better utilization. An energy efficient task scheduling based on QoS parameter has been proposed to address above said challenge. Firsty, all the incoming tasks are categorized into four classes based on some special attributes and prioritize according to importance of the classes. Secondly, Physical Machines (PMs) type confirmation list is selected based on the number of resource blocks and then select one PM that has maximum QoS value. All the Virtual Machines (VMs) on selected PM are prioritized according to their weight. Experimental evaluation done on CloudSim shows the effectiveness and efficiency of proposed approach
Augmented MRI Images for Classification of Normal and Tumors Brain through Transfer Learning Techniques
Get PDFA brain tumor is a severe malignant condition caused by uncontrolled and abnormal cell division. Recent advances in deep learning have aided the health business in Medical Imaging for the diagnosis of numerous disorders. The most frequent and widely used deep learning algorithm for visual learning and image recognition. This research seeks to multi-classification tumors in the brain from images attained by Magnetic Resonance Imaging (MRI) using deep learning models that have been pre-trained for transfer learning. As per the publicly available MRI brain tumor dataset, brain tumors identified as glioma, meningioma, and pituitary, are accounting for most brain tumors. To ensure the robustness of the suggested method, data acquisition, and preprocessing are performed in the first step followed by data augmentation. Finally, Transfer Learning algorithms including DenseNet, ResNetV2, and InceptionResNetv2 have been applied to find out the optimum algorithm based on various parameters including accuracy, precision, and recall, and are under the curve (AUC). The experimental outcomes show that the model’s validation accuracy is high for DenseNet (about 97%), while ResNetv2 and InceptionResNetv2 achieved 77% and 80% only
Keyless Security: The Smart Solution for Home with a Smart Door Lock
Get PDFThis research paper introduces a novel approach to improve security and access control in smart door lock systems. The proposed system incorporates multi-factor authentication, combining a numerical keypad, mobile app, and RFID sensor. Each authentication method is associated with a specific lock, offering differentiated access control. Notably, the system requires the simultaneous use of the mobile app and RFID sensor to unlock a designated lock, adding an extra layer of security. To assess the system's effectiveness, a comprehensive study was conducted, considering security, usability, and user satisfaction. The study involved simulated attacks, user surveys, and usability tests, with data analysis providing insights into the system's resilience and user acceptance. This research contributes to the field of smart door lock systems by offering an innovative approach to enhance security through multi-factor authentication. The findings provide valuable insights for designing advanced access control mechanisms, empowering users with increased confidence in their security measures while maintaining a user-friendly experience
Heart Disease Prediction Using Ensemble Model and Hyperparameter Optimization
Get PDFHeart disease is a major global health concern that responsible for significant mortality rates, killing 17.9 million people each year on average. To overcome this problem, machine learning can assist in forecasting the occurrence of heart disease, aiding in its prevention and treatment. This paper explores several classification models to forecast heart disease. This paper also utilizes the hyperparameter tuning method via grid search cv to enhance the accuracy of the models. Finally, the experiment concludes with an ensemble vote on all hyperparameter-tuned classification models. The x-gradient boost and random forest classifier deliver the best outcomes, with an accuracy of 88.04% and 89.13% before hyperparameter optimization, and 92.39% after hyperparameter optimization. These results show that machine learning models are capable of forecasting the risk of heart disease. These models may assist healthcare professionals in identifying individuals at risk of heart disease, enabling preventative measures to be taken. It is essential to note that this study focuses solely on classification models and may not represent the entire population. Further research is required to determine the predictability of heart disease in diverse populations
Classification of Acute Lymphocytic Leukemic Blood Cell Images using Hybrid CNN-Enhanced Ensemble SVM Models and Machine Learning Classifiers
Get PDFAcute Lymphocytic Leukemia is a dangerous kind of malignant cancer caused due to the overproduction of white blood cells. The white blood cells in our body are responsible for fighting against infections, if the WBC increases the immunity will decrease and it would lead to serious health conditions. Malignant cancers such as ALL is life threatening if the disease is not diagnosed at an early stage. If a person is suffering from ALL the disease needs to be diagnosed at an early stage before it starts spreading, if it starts spreading the person’s chances of survival would also reduce. Here comes the need of an accurate automated system which would assist the oncologists to diagnose the disease as early as possible. In this paper some of the algorithms that are enhanced to detect and classify ALL are incorporated. In order to classify the Acute Lymphocytic Leukemia a hybrid model has been deployed to improve the accuracy of the diagnosis and it is termed as Hybrid CNN Enhanced Ensemble SVM for the classification of malignancy. Machine Learning classifiers are also used to design the system and it is then compared with enhanced CNN based on the performance metrics
Attack Classification and Detection for Misbehaving Vehicles using ML/DL
Get PDFVehicle ad hoc networks are a crucial component of the next Intelligent Transportation System created to build a reliable and secure connection between various network components to establish a safe and effective transportation network. Because of open nature of VANETs become vulnerable to numerous assaults such forgery, Denial-of-Service (DoS), and false reports, which can ultimately cause traffic jams or accidents The earlier study concentrated on misbehaving vehicles rather than RSUs. Proposed method integrates data from two subsequent BSMs for testing and training by employing machine learning (ML) methods. The framework merges the data from two BSMs in the right manner and utilizes machine learning/Deep learning methodology which identify the running vehicle as a legal or hostile one