Asian Journal of Convergence in Technology
Not a member yet
868 research outputs found
Sort by
Motorcycle Apprehension using Deep Learning and K-Nearest Neighbor Algorithm
Get PDFAbstract — Road violations that lead to accidents and deaths are increasing significantly. There are about 1.35 million people who die every year because of road accidents, and more than half of these involve a motorcycle. Authorities are strictly implementing traffic laws and making some innovations to capture those motorists violating laws easily. Researchers are also doing their part to help solve the problem; indeed, their studies give a vast contribution and solve road safety issues. However, the papers on road violations were focused more on on-road violations involving four-wheeled vehicles.
For this reason, a motorcyclist violation detection and plate recognition with e-mail notification using a Deep Learning algorithm were developed to apprehend motorcyclists violating traffic laws. Tensorflow Object Detection API was used as a framework along with the Faster R-CNN model. The system was developed using Anaconda Environment, Python Scripting, KNN, and MySQL Connector. The conditions and criteria for detecting a violation are based on motorcycle detection, including motorcycle tracking. After violation detection and plate recognition, the violation's image is sent through e-mail together with the details of the offense
Knowledge Management framework for the supervision of IT postgraduate research in Sri Lanka
Get PDFThe private sector higher education industry is increasingly attracting a knowledge-based community that depends critically on Knowledge Management (KM) and Knowledge Sharing (KS) activities to expand the quality of supervising postgraduate research students. Using the KM approach to share good research supervision knowledge will help junior research supervisors to conduct quality research with students and thereby help the supervision process to be more successful. The objective of this study is to suggest a conceptual framework that fits in the supervision process. This is conducted to investigate how KM and Information Technology (IT) can be used to develop a model for the supervision process. The framework highlights the critical KM activities in the research supervision process, and it is based on the Task/Technology Fit theory. Using this framework, the knowledge of the more experienced supervisors will be captured and used by junior supervisors in their supervision process
IOT Based Application for monitoring and Predicting Air Quality in the environment
Get PDFIndia has been dealing with pollution for a long time. As the matter of fact, in February, India was home to six out of 10 of the world’s most polluted cities. While on the contrary, atmospheric pollution, surrounding pollutants are about 1000 times more likely to be transmitted to the lungs, causing diseases The main sources of air pollution are motor vehicle emissions, illegal industrial activities, harmful pesticides, and many times we see LPG gas leakages and cylinder truck accidents on the road which are harmful to us and can take our lives. Therefore, poor air quality results in many health complications, such as heart disease, lung cancer, and breathing problems like asthma. It is important to regulate air pollution and also to incorporate technology, sensors, and software systems to ensure that air pollution is closely monitored. Our efforts in this project are to create an application that can be used to track air quality to take preventive steps to keep our living environment healthy. The program is user-friendly and works to produce real-time air quality warnings as a preventive mechanism. Our key contribution is to build a monitoring system for air quality that senses the real-time data of surrounding parameters such as carbon monoxide and PM level and warns people when the sum of these elements goes above a certain limit and presents the data in a format that is easy to understand
Ocular Disease Detection Using Advanced Neural Network Based Classification Algorithms
Get PDFOne of the most challenging tasks for ophthalmologists is early screening and diagnosis of ocular diseases from fundus images. However, manual diagnosis of ocular diseases is difficult, time-consuming and it can be prone to errors. That is why a computer-aided automated ocular disease detection system is required for the early detection of various ocular diseases using fundus images. Due to the enhanced image classification capabilities of deep learning algorithms, such a system can finally be realized. In this study, we present four deep learning-based models for targeted ocular tumor detection. For this study, we trained the cutting-edge image classification algorithms such as Resnet-34, EfficientNet, MobileNetV2, and VGG-16 on the ODIR dataset consisting of 5000 fundus images that belong to 8 different classes. Each of these classes represents a different ocular disease. The VGG-16 model achieved an accuracy of 97.23%; the Resnet-34 model reached an accuracy of 90.85%; the MobileNetV2 model provided an accuracy of 94.32%, and the EfficientNet classification model achieved an accuracy of 93.82%. All of these models will be instrumental in building a real-time ocular disease diagnosis system
I-V Characteristics of Cadmium Telluride Quantum Dots Diode Fabricated by Drop Casting Method
Get PDFCurrent (I)-Voltage (V) characterizations of a CdTe quantum dot diode is discussed in this article. The quantum dot diode is fabricated by a simple drop casting method on an n-type Indium Tin Oxide glass substrate. The linear variation of current with respect to applied voltage can be attributed to reduced grain boundary defects and enriched crystallinity. It is also observed that quantum dots can show good current conduction than CdTe nano-rods
Social Impact on Hybrid and Retrofitted Three-Wheelers
Get PDFTo boost electric vehicles on Indian roads government brought up a policy that can eliminate personal vehicles in the upcoming decades. With an increase in shared mobility such as metros, monorails, and BRT systems, last-mile connectivity has already acquired the importance of never before. Bicycles and electric three-wheelers can fit this sector. Electrically operated three-wheelers will also gain importance in last-mile goods delivery. Similarly, the Internal Combustion Engine (ICE) based three-wheelers will pass through a phase of retrofitting and use of hybrid energy sources to fulfill economic and technological needs. It is very important to study retrofitting and hybrid three-wheelers in detail to understand how they can fill the gap of transition from ICE based system to a fully electrically operated model. These vehicles will have a social impact on consumers as they are cost-effective vehicles. Innovative, swapping of battery and fuel tank is proposed to address some of the needs in this transition. This paper presents techno-commercial aspects of retrofitted three-wheelers and how to bring an economically viable hybrid version of three-wheelers
Electrical vehicle speed control by AI technique
Get PDFThis research paper in regards to the programmed vehicle driving framework without driver. This part presents a speed control of electric vehicle (EV) to improve the comportment and solidness under different street essential state . The control circuit utilizing shrewd fuzzy PI regulator is proposed. Boundaries which manage the working of PI regulator are progressively changed with the help of fuzzy canny control. This work manages the issue of way following for an electrical vehicle which has four electromechanical wheels framework. A regulator dependent on fuzzy rationale send the speed and the guiding guidelines to the robot to guarantee that the robot follows the direction , during the plan of this regulator the exertion was focused on the effortlessness and proficiency . The handling and reaction time are fundamental components in the process control, which is the thing that roused the decision of the construction and the boundaries of the various pieces of the fuzzy regulator
Blockchain for an Alternative GPS
Get PDFGlobal Positioning System is very critical for many applications. If it is out of service, there may be chaotic situations for the applications. For this reason, there should be other types of sources for location information. In this work, a blockchain is proposed for location information. Blockchain provides a resilient system, and it can also provide reliable location information. Reliability of location information is also supported by short-range communication. Short-range communication eliminates location information errors outside the communication range. Therefore, it helps to minimize location information errors. In the blockchain, there are special devices to provide location information. The proposed blockchain is a market for location trade with its own cryptocurrency, which is used mostly to incentivize the devices to share location information among themselves. Moreover, the proposed blockchain respect location privacy using encryption mechanism
AI based Adaptive Network for Smart Cities
Get PDFSelf-aware, Self-Defending Adaptive Network is a network that defends itself from security breaches in the deployment of smart cities. An adaptive system that knows and recognizes the level of threat faced by an intrusion across a network of smart cities. To detect and separate security threats, the AI program uses a new method of machine learning to track any aspect of a network. Threats include ransomware, high-jacking code, intrusion and illegal entry, theft, and unauthorized use. An autonomous system comprises of a collection of autonomous modules that are introduced and removed dynamically. To order to achieve machine objectives, nodes inside such an ensemble will cooperate. In response to changes in its operating environment, the self-adaptive network modifies its own behavior. We mean anything that the network can observe, such as user interaction, network devices and sensors, or instrumentation, by operating environment
Full Time Result Prediction using Ensemble Techniques
Get PDFSports Analytics is a growing industry and one of the best real-word applications of Data Science. In this paper, the interest of author and machine learning capabilities were combined to develop a result predictor for football matches. The model proposed is capable of predicting result of any English Premier League Match at the half-time with 75% accuracy. The full-time result predictor is a system based on ensemble of powerful classification algorithms which can predict the odds of winning and draw of both home team and away team on the basis of goals scored at the half time and the current standings in the league. The model learns from the past records of the league and the results of different models are compared in the last section of the paper