Asian Journal of Convergence in Technology
Not a member yet
868 research outputs found
Sort by
HANDLING BIG TABULAR DATA OF ICT SUPPLY CHAINS: A MULTI-TASK, MACHINE-INTERPRETABLE APPROACH
Get PDFThe essential details of ICT devices are frequently distilled into large tabular data sets that are distributed throughout supply chains as a result of the features of Information and Communications Technology (ICT) goods. With the explosion of electronic assets, it is crucial to automatically analyse tabular structures. We develop a Table Structure Recognition (TSR) work and a Table Cell Type Classification (CTC) task to convert the tabular data in electronic documents into a machine-interpretable format and give layout and semantic information for information extraction and interpretation. For the TSR job, complicated table structures are represented using a graph. Table cells are divided into three groups—Header, Attribute, and Data—based on how they work for the CTC job. Then, utilising the text modal and picture modal characteristics, we provide a multi-task model to accomplish the two tasks concurrently. Our test findings demonstrate that, using the ICDAR2013 and UNLV datasets, our suggested strategy can beat cutting-edge approaches
Development of an IOT Based Industrial Fault Detection and Diagnosis System
Get PDFThis paper has proposed an IoT based industrial fault detection and diagnosis system. The IoT (Internet of things) based industry protection system employing Arduino is intended to safeguard industries from losses brought on by accidents and monitoring the faults. Industrial accidents such as gas leakage causes fire resulting in significant industrial setbacks. Due to furnace explosions, electrical short circuits, or other circumstances, quick fire detection is also required. The proposed system uses flame sensor along with gas sensor to detect fire as well as gas leakage to avoid any industrial mishaps and prevent economic damage. The system consists of temperature detector to detect the high/low temperature, the phase fault detector (either single or three phases), voltage controller (high/low), and frequency monitoring. Flame and Gas sensors are also interfaced with arduino and LCD screen. The sensor data is constantly scanned to record values and check for fire, gas leakage and then this data is transmitted to online. The wifi module is used to achieve internet functionality. GSM module is sent message to specific numbers in case human operators aren’t present in online always. IP camera monitor the whole system from anywhere in the world
Fault Diagnosis of a Transformer using Fuzzy Model and Unsupervised Learning
Get PDFIn this paper a power transformer fault diagnosis system (PDFDS) based on soft computing and computational intelligence is proposed. Fault diagnosis and analysis is an integral part of operational reliability. Systems like SCADA collect data of various equipment in power system network, however, fails to provide a critique fault diagnosis for the same which further leads to additional cost of replacing the equipment. This paper proposes a supervised-unsupervised predictive model for the data collected from various power transformers across Himachal Pradesh and IEC 10 database. To identify the different fault types in a transformer a fuzzy model is developed using the DGA interpretation techniques. Since, not all data samples in the collected dataset fall under the standards specified in the ratio tables it thus becomes difficult to identify the type of fault for such cases. To overcome this an improved fuzzy model with unsupervised clustering algorithm or Fuzzy Clustering means is used. Employing this improved model optimizes the data before feeding it to the different predictive machine learning models. Further, a particle swarm optimization algorithm with passive congregation is employed to optimize the performance of these machine learning models
Structural Stress Analysis for Cornering Fatigue Test of Wheel Rim as per AIS 073 using Finite Element Method
Get PDFAll vehicle components are undergone with the physical testing to certify the safety standard and to designvalidation. Experimental testing is taking too much time and sothe overall design process is getting slow. Now a days to minimize the design cycle time most of suppliers and OEM are perform the FE simulation. FE simulation is commonly used for the full vehicle analysis and component level. FEA simulation can be able to provide the results with minimum % error. In this project I have perform the finite element analysis of the cornering fatigue test for the E vehicle automobile wheelrim as per the automotive industry standard AIS 073. The objective of this analysis is to find out the critical and high stress concentration region for the CFT loading for further design modification and to minimize the design cycle time
Research on RFID-Based Wallet Detection System
Get PDFMany of us mostly forget our Wallet, which is a must for all as it allows for the safeguarding very priceless things. Wallet is one of the most belongings, We are thus implementing a Smart Almost following you wallet. When compared to the standard wallets that are sold in the market, the smart wallet has several advantages. It is a contemporary wallet that has a GPS system integrated into it. Simply connect your smartphone to your smart wallet and use the app to track it. There are two operating modes for the Smart Wallet: Normal mode and Lost mode. The Wallet goes into lost mode when it is far away. The misplaced wallet can then be located using a GPS system. While in the default mode, if we travel a particular distance away from our wallet, we receive an alarm or alert message that we neglected to bring our wallet, allowing us to quickly locate it and carry it with us. The smart wallets can do all the tasks that the traditional dumb wallet can, but it also has additional capabilities like mobile charging and anti- theft security. It is always more expensive than a typical wallet, but it is typically well-made, has a sleek appearance, and offers some additional technological incentives. In order to address this, the research report suggests a smart wallet with features including a notification system for users who move away from the wallet
Novel Approach with Deep Learning Models For Melanoma Skin Cancer Detection
Get PDFArtificial intelligence methodologies with deep learning models proved very effective in many areas. This research focuses on using deep learning models for the detection of skin cancer melanoma in patients at an early stage. The growth of a malignant melanocytic tumor is the primary cause of melanoma, a deadly form of skin cancer. The most dangerous type of cancer, known as melanoma, is brought on by melanocytes, which produce pigment. Seventy-five percent of skin cancer fatalities are caused by melanoma. Only 15% of patients who have survived over the previous five years, according to a review of the survival rate, have received chronic treatment. The manual system is the issue. The mole has a maximum of six colors, thus the image is troublesome since it has a variety of tints and is hard for humans to distinguish. As a result, we created a model utilizing flask and a pre-trained deep-learning model to handle this issue. The suggested model is optimized using the SGD, RMSprop, Adam, Agagrad, Adadelta, and Nadam optimizers. According to testing, the suggested CNN model with the Adam optimizer performs the best at categorizing the dataset of skin cancer lesions. Also it gives the result specifically to the seven types of the skin cancer with their possible chances in percentage. It achieved a training accuracy of 99.73 % and a testing accuracy of 96.53, which is better than the previous results. In order to lower the mortality rate, this research intends to identify an early treatment for people with skin cancer
A Survey on Waste Water Treatment and Cleaning Using Numerous Techniques
Get PDFThis survey paper presents various sustainable methods to remove moss from tributaries, which can be profitable under certain conditions. Moss can interfere with wastewater treatment and contribute to the pollution of surface water downstream, including heavy metal contamination. The paper introduces a robotic moss remover called ARROS, which operates with lower power consumption and less workload compared to traditional methods. ARROS is a catamaran type unmanned surface vehicle (USV) equipped with guidance, navigation, and control (GNC) equipment and harmful Moss blooms (HABs) tools with electrocoagulation and flotation (ECF) technology. The USV communicates with an unmanned aerial vehicle (UAV) server to detect algal blooms accurately, and image and texture-based recognition algorithms can identify HABs and send their location to the USV for route planning. The paper proposes to use Failure Mode and Effects Analysis (FMEA) processing technology to analyze the wet cleaning system in this study. Additionally, the paper highlights the filamentous Moss structure composed of Moss lawn cleaner and filamentous Moss nutrition cleaner, which have received little attention in the past 15 years. Overall, this survey paper reviews various Moss cultivation techniques and introduces a unique robotic Moss remover that can help prevent future inventions from failing. The keywords of the paper include Moss scrubbers, Algal bloom, filamentous Moss nutrient scrubbers (FANS), Harmful algal blooms (HABs), Electrocoagulation and flotation (ECF), and Image-based moss bloom detectio
Automating Machinery with Object Detection using YOLO and Servo Controllers
Get PDFNow-a-days Computer Vision and Machine Learning algorithms play an important role in automation. With the help of Computer Vision and deep learning algorithms, data like images and videos are being used for classification and prediction. This paper proposes a real time object detector using computer vision and deep learning algorithms. YOLO (You Only Look Once) which is a deep learning algorithm is a state-of-the-art algorithm used for object detection. A binary classifier using CNN (Convolutional Neural Network) can be used to detect whether a class is present or absent indicating the presence or absence of a particular object. The two classes will be A) desired object present and B) the desired object absent. The input to the model will be from a live camera. The classifier detects the object by creating a bounding box around the object and then predicting the class. If class A is predicted the model will calculate the distance from the object. After calculating the distance, the model will instruct the machine to pick up the object and place it on the required position. If class ‘B’ is present the model will just display the message stating the class is absent
IRIS RECOGNITION USING IMAGE PROCESSING
Get PDFIn the contemporary landscape of digital evaluations, ensuring robust and accurate user verification is paramount. Traditional methods like passwords and PINs are prone to numerous vulnerabilities such as phishing and brute force attacks. Addressing these concerns, iris recognition emerges as a formidable biometric authentication solution. This research proposes the incorporation of iris recognition technology into an online assessment framework leveraging Java Server Pages (JSP). By utilizing iris recognition to verify users before granting access to assessment content, this innovative system enhances security protocols and reduces the likelihood of unauthorized access.
The system's architecture leverages JSP for dynamic web page generation, seamlessly integrating iris capture and recognition functionalities into the web interface. Key components of the system encompass iris image capture, feature extraction, matching, and user authentication. Iris images can be obtained through webcam interactions or file uploads, with extracted features compared against secure database templates for user authentication. By integrating iris recognition logic with JSP, real-time authentication is achieved during user login and assessment sessions.
The system's implementation adheres strictly to industry best practices in web development, ensuring scalability, reliability, and cross-platform compatibility. Addressing privacy and security concerns surrounding biometric data, encryption techniques are employed, and regulatory standards are strictly followed. Ultimately, this system drives advancements in online assessment security by harnessing biometric technology within the familiar framework of JSP-based web applications
ChainBank: Empowering Students to Build Strong Financial Habits
Get PDFChainBank is an innovative decentralized autonomous organizational system specifically designed to support students in improving their financial habits. Students have the opportunity to invest their money in a distributed pool and receive tokens in return, while also gaining the ability to actively engage in decision-making through voting on proposals. ChainBank offers its members a powerful voting feature, enabling them to collectively make important decisions regarding the platform’s operations. ChainBank’s primary objective is to target regular students who often struggle with irregular financial habits. Engaging with ChainBank helps students to develop essential skills and knowledge that will positively impact their financial well-being both now and in the future