International Journal on Future Revolution in Computer Science & Communication Engineering
Not a member yet
1384 research outputs found
Sort by
Visual Cryptography-Based Secure QR Payment System Design and Implementation
Get PDFIt is important to validate the Merchant and the Client to increase confidence in online transactions. At present, only the Client is checked against the merchant server. The research in this paper will show you how to create and launch a QR code-based payment system that is both secure and convenient for users. As a result of their capacity to facilitate instantaneous transactions and offer unparalleled ease of use, QR codes have seen explosive growth in the past few years. QR-based online payment systems are easy to use but susceptible to various assaults. So, for the level of security given by transaction processing to hold, the secrecy and integrity of each payment procedure must be guaranteed. In addition, the online payment system must verify each transaction from both the sender's and the recipient's perspectives. The study's QR-based method is kept safe through visual cryptography. The suggested approach takes advantage of visual cryptography via a web-based application
The Study and Efficacy of Conventional Machine Learning Strategies for Predicting Cardiovascular Disease
Get PDFRegarding medical science, cardiovascular disease is the main cause of death. Testing patient samples for cardiac disease can save lives and lower mortality rates. During a subsequent visit, the right remedies should be outlined and prescribed. One of the most important factors in preemptive cardiac disease diagnosis is accuracy. Based on this factor, many research approaches were examined and compared. According to the analysis of these approaches, new procedures appear to be more advanced and reliable in detecting cardiac illness. A notation of the methods and their underlying themes and precision levels will be discussed. This paper surveys many models that use these methods and methodologies and evaluates their performance. Models created utilizing supervised learning methods, such as Support Vector Machines (SVM), K-Nearest Neighbour (KNN), Decision Trees (DT), Random Forest (RF), and Logistic Regression Units, are highly valued by researchers. For benchmark datasets like the Cleveland or Kaggle, the methodologies are derived from data mining, machine learning, deep learning, and other related techniques and technologies. The accuracy of the provided methods is graphically demonstrated
IOT based Security System for Auto Identifying Unlawful Activities using Biometric and Aadhar Card
Get PDFIn today’s era, where thefts are consecutively increasing, especially in banks, jewelry shops, stores, ATMs, etc, there is a need to either develop a new system or to improve the existing system, due to which the security in these areas can be enhanced. However, the traditional methods (CCTV cameras, alarm buttons) to handle the security issues in these areas are still available, but they have lots of limitations and drawbacks. So, in order to handle the security issues, this paper describes how the biometric and IoT (Internet of Things) techniques can greatly improve the existing traditional security system. Our proposed system uses biometric authentication using the fingerprint and iris pattern with the strength of IoT sensors, microcontroller and UIDAI aadhar server to enhance the security model and to cut the need of keeping extra employees in monitoring the security system
Numerical Modeling and Design of Machine Learning Based Paddy Leaf Disease Detection System for Agricultural Applications
Get PDFIn order to satisfy the insatiable need for ever more bountiful harvests on the global market, the majority of countries deploy cutting-edge technologies to increase agricultural output. Only the most cutting-edge technologies can ensure an appropriate pace of food production. Abiotic stress factors that can affect plants at any stage of development include insects, diseases, drought, nutrient deficiencies, and weeds. On the amount and quality of agricultural production, this has a minimal effect. Identification of plant diseases is therefore essential but challenging and complicated. Paddy leaves must thus be closely watched in order to assess their health and look for disease symptoms. The productivity and production of the post-harvest period are significantly impacted by these illnesses. To gauge the severity of plant disease in the past, only visual examination (bare eye observation) methods have been employed. The skill of the analyst doing this analysis is essential to the caliber of the outcomes. Due to the large growing area and need for ongoing human monitoring, visual crop inspection takes a long time. Therefore, a system is required to replace human inspection. In order to identify the kind and severity of plant disease, image processing techniques are used in agriculture. This dissertation goes into great length regarding the many ailments that may be detected in rice fields using image processing. Identification and classification of the four rice plant diseases bacterial blight, sheath rot, blast, and brown spot are important to enhance yield. The other communicable diseases, such as stem rot, leaf scald, red stripe, and false smut, are not discussed in this paper. Despite the increased accuracy they offer, the categorization and optimization strategies utilized in this work lead it to take longer than typical to finish. It was evident that employing SVM techniques enabled superior performance results, but at a cost of substantial effort. K-means clustering is used in this paper segmentation process, which makes figuring out the cluster size, or K-value, more challenging. This clustering method operates best when used with images that are comparable in size and brightness. However, when the images have complicated sizes and intensity values, clustering is not particularly effective
Analysis and Design of Detection for Liver Cancer using Particle Swarm Optimization and Decision Tree
Get PDFLiver cancer is taken as a major cause of death all over the world. According to WHO (World Health Organization) every year 9.6 million peoples are died due to cancer worldwide. It is one of the eighth most leading causes of death in women and fifth in men as reported by the American Cancer Society. The number of death rate due to cancer is projected to increase by45 percent in between 2008 to 2030. The most common cancers are lung, breast, and liver, colorectal. Approximately 7, 82,000 peoples are died due to liver cancer each year. The most efficient way to decrease the death rate cause of liver cancer is to treat the diseases in the initial stage. Early treatment depends upon the early diagnosis, which depends on reliable diagnosis methods. CT imaging is one of the most common and important technique and it acts as an imaging tool for evaluating the patients with intuition of liver cancer. The diagnosis of liver cancer has historically been made manually by a skilled radiologist, who relied on their expertise and personal judgement to reach a conclusion. The main objective of this paper is to develop the automatic methods based on machine learning approach for accurate detection of liver cancer in order to help radiologists in the clinical practice. The paper primary contribution to the process of liver cancer lesion classification and automatic detection for clinical diagnosis. For the purpose of detecting liver cancer lesions, the best approaches based on PSO and DPSO have been given. With the help of the C4.5 decision tree classifier, wavelet-based statistical and morphological features were retrieved and categorised
Image Processing-Based Lung Cancer Detection Using Adaptive CNN Mixed Sine Cosine Crow Search Algorithm in Medical Applications
Get PDFMedical image processing relies heavily on the diagnosis of lung cancer images. It aids doctors in determining the correct diagnosis and management. For many patients, lung cancer ranks among the most deadly diseases. Many lives can be saved if cancerous growth is diagnosed early. Computed Tomography (CT) is a critical diagnostic technique for lung cancer. There was also an issue with finding lung cancer due to the time constraints in using the various diagnostic methods. In this study, an Adaptive CNN Mixed Sine Cosine Crow Search (ACNN-SCCS) strategy is proposed to assess the presence of lung cancer in CT images based on the imaging technique. Accordingly, the presented classification scheme is used to assess these traits and determine whether or not the samples include cancerous cells. To obtain the highest level of accuracy for our research the proposed technique is analyzed and compared to many other approaches, and its performance metrics (detection accuracy, precision, f1-score, recall, and root-mean-squared error) are examined
Mobile Cloud IoT for Resource Allocation with Scheduling in Device- Device Communication and Optimization based on 5G Networks
Get PDFInternet of Things (IoT) is revolutionising technical environment of traditional methods as well as has applications in smart cities, smart industries, etc. Additionally, IoT enabled models' application areas are resource-constrained as well as demand quick answers, low latencies, and high bandwidth, all of which are outside of their capabilities. The above-mentioned issues are addressed by cloud computing (CC), which is viewed as a resource-rich solution. However, excessive latency of CC prevents it from being practical. The performance of IoT-based smart systems suffers from longer delay. CC is an affordable, emergent dispersed computing pattern that features extensive assembly of diverse autonomous methods. This research propose novel technique resource allocation and task scheduling for device-device communication in mobile Cloud IoT environment based on 5G networks. Here the resource allocation has been carried out using virtual machine based markov model infused wavelength division multiplexing. Task scheduling is carried out using meta-heuristic moath flame optimization with chaotic maps. So, by scheduling tasks in a smaller search space, system resources are conserved. We run simulation tests on benchmark issues and real-world situations to confirm the effectiveness of our suggested approach. The parameters measured here are resource utilization of 95%, response time of 89%, computational cost of 35%, power consumption of 38%, QoS of 85%
MMwave MIMO in 5G Network Analysis for Spectral Efficiency with Beamforming Based Channel Estimation
Get PDF5G network has its high energy efficiency and spectrum efficiency, massive multiple-input and multiple-output (MIMO) has been envisioned as a key technology.This research work is centred on optimal method creation of energy-efficient massive MIMO methods, which is most active research technology in the communication industry.The suggested model, which takes into account a multi-cell model scenario, is a realistic method that improved spectral efficiency (SE) of huge MIMO methods.Base stations (BSs) do channel estimate based on uplink (UL) transmission using least-square (LS), element-wise MMSE, and minimum mean-squared error (MMSE) estimators.This research propose novel technique in MMwaveMIMO 5G network based spectral efficiency and channel estimation. The aim of this research is to enhance the spectral efficiency of MIMO channel using HetNets zero forcing Multiuser propagation models. The channel estimation is carried out based on beamforming using matched filter channel estimation with wide band antenna.Finally, simulation results demonstrate the high channel estimate accuracy and spectrum efficiency that the suggested systems can accomplish.Proposed technique attained sum rate of 85%, spectral efficiency of 93%, DoF of 79%, energy efficiency of 98% and detection accuracy of 96% for number of cells and sum rate of 77%, spectral efficiency of 85%, DoF of 71%, energy efficiency of 92% and detection accuracy of 95% for number of users
Underwater Aerial Vehicle Networks Based Image Analysis By Deep Learning Architecture Integrated With 5G System
Get PDFWith its astonishing ability to learn representation from data, deep neural networks (DNNs) have made efficient advances in the processing of pictures, time series, spoken language, audio, video, and many other types of data.In an effort to compile the volume of information generated in remote sensing field's subfields, surveys and literature revisions explicitly concerning DNNs methods applications are carried out Aerial sensing research has recently been dominated by applications based on Unmanned Aerial Vehicles (UAVs).There hasn't yet been a literature review that integrates the "deep learning" and "UAV remote sensing" thematics.This research propose novel technique in underwater aerial vehicle networks based image analysis by feature extraction and classification utilizing DL methods. here UAV based images through on 5G module is collected and this image has been processed for noise removal, smoothening and normalization. The processed image features has been extracted using multilayer extreme learning based convolutional neural networks. Then extracted deep features has been classified utilizingrecursive elimination based radial basis function networks. The experimental analysis is carried out based on numerous UAV image dataset in terms of accuracy, precision, recall, F-measure, RMSE and MAP.Proposed method attained accuracy of 96%, precision of 94%, recall of 85%, F- measure of 72%, RMSE of 48%, MAP of 41%
5G Network in Content Based Emotion Detection by Sentimental Analysis Integrated with Opinion Mining and Deep Learning Architectures
Get PDFThe rapid growth of social networking sites in the Internet era has made them a necessary tool for sharing emotions with the entire world. To extract emotions from text, a variety of tools and approaches are available in fields of opinion mining as well as sentiment analysis. These researches propose novel technique opinion mining based emotion detection from the input social content using deep learning architectures. Here the input has been obtained as social media content based on opinion miningby 5G networks. The input has been processed for noise removal, smoothening and normalization. This processed input has been segmented using Markov model based convolutional neural networks (MMCNN). The segmented data has been classified using Canonical Correlation AnalysisBayesian neural network.An opinion mining method that analyzes statements regarding computer programming and predicts or recognizes their polarity was implemented, along with an earlier module that was integrated into an intelligent learning environment. These three steps made up the creation of the module. We assessed the corpus, text polarity precision, and emotion recognition. Experimental analysis has been carried out for various social media content collected by opinion mining in terms of accuracy, precision, recall, F-1 score, AUC.Proposed technique attained accuracy of 99%, precision of 96%, recall of 96%, F-1 score of 95%, AUC of 89%