World Academy of Science, Engineering and Technology
[Computer and Information Engineering]
Online ISSN : 1307-6892
3053 Detecting Characters as Objects Towards Character Recognition on Licence Plates
Authors: Alden Boby, Dane Brown, James Connan
Abstract:
Character recognition is a well-researched topic across disciplines. Regardless, creating a solution that can cater to multiple situations is still challenging. Vehicle licence plates lack an international standard, meaning that different countries and regions have their own licence plate format. A problem that arises from this is that the typefaces and designs from different regions make it difficult to create a solution that can cater to a wide range of licence plates. The main issue concerning detection is the character recognition stage. This paper aims to create an object detection-based character recognition model trained on a custom dataset that consists of typefaces of licence plates from various regions. Given that characters have featured consistently maintained across an array of fonts, YOLO can be trained to recognise characters based on these features, which may provide better performance than OCR methods such as Tesseract OCR.Keywords: computer vision, character recognition, licence plate recognition, object detection
Procedia PDF Downloads 1123052 Identification of the Relationship Between Signals in Continuous Monitoring of Production Systems
Authors: Maciej Zaręba, Sławomir Lasota
Abstract:
Understanding the dependencies between the input signal, that controls the production system and signals, that capture its output, is of a great importance in intelligent systems. The method for identification of the relationship between signals in continuous monitoring of production systems is described in the paper. The method discovers the correlation between changes in the states derived from input signals and resulting changes in the states of output signals of the production system. The method is able to handle system inertia, which determines the time shift of the relationship between the input and output.Keywords: manufacturing operation management, signal relationship, continuous monitoring, production systems
Procedia PDF Downloads 863051 Analysis and Prediction of Netflix Viewing History Using Netflixlatte as an Enriched Real Data Pool
Authors: Amir Mabhout, Toktam Ghafarian, Amirhossein Farzin, Zahra Makki, Sajjad Alizadeh, Amirhossein Ghavi
Abstract:
The high number of Netflix subscribers makes it attractive for data scientists to extract valuable knowledge from the viewers' behavioural analyses. This paper presents a set of statistical insights into viewers' viewing history. After that, a deep learning model is used to predict the future watching behaviour of the users based on previous watching history within the Netflixlatte data pool. Netflixlatte in an aggregated and anonymized data pool of 320 Netflix viewers with a length 250 000 data points recorded between 2008-2022. We observe insightful correlations between the distribution of viewing time and the COVID-19 pandemic outbreak. The presented deep learning model predicts future movie and TV series viewing habits with an average loss of 0.175.Keywords: data analysis, deep learning, LSTM neural network, netflix
Procedia PDF Downloads 2303050 Knowledge and Ontology Engineering in Continuous Monitoring of Production Systems
Authors: Maciej Zaręba, Sławomir Lasota
Abstract:
The monitoring of manufacturing processes is an important issue in nowadays ERP systems. The identification and analysis of appropriate data for the units that take part in the production process are ones of the most crucial problems. In this paper, the authors introduce a new approach towards modelling the relation between production units, signals, and factors possible to obtain from the production system. The main idea for the system is based on the ontology of production units.Keywords: manufacturing operation management, OWL, ontology implementation, ontology modeling
Procedia PDF Downloads 1083049 The Relationship between Human Pose and Intention to Fire a Handgun
Authors: Joshua van Staden, Dane Brown, Karen Bradshaw
Abstract:
Gun violence is a significant problem in modern-day society. Early detection of carried handguns through closed-circuit television (CCTV) can aid in preventing potential gun violence. However, CCTV operators have a limited attention span. Machine learning approaches to automating the detection of dangerous gun carriers provide a way to aid CCTV operators in identifying these individuals. This study provides insight into the relationship between human key points extracted using human pose estimation (HPE) and their intention to fire a weapon. We examine the feature importance of each keypoint and their correlations. We use principal component analysis (PCA) to reduce the feature space and optimize detection. Finally, we run a set of classifiers to determine what form of classifier performs well on this data. We find that hips, shoulders, and knees tend to be crucial aspects of the human pose when making these predictions. Furthermore, the horizontal position plays a larger role than the vertical position. Of the 66 key points, nine principal components could be used to make nonlinear classifications with 86% accuracy. Furthermore, linear classifications could be done with 85% accuracy, showing that there is a degree of linearity in the data.Keywords: feature engineering, human pose, machine learning, security
Procedia PDF Downloads 833048 Study of a Decentralized Electricity Market on Awaji Island
Authors: Arkadiusz P. Wójcik, Tetsuya Sato, Shin-Ichiro Shima, Mateusz Malanowski
Abstract:
Over the last decades, new technologies have significantly changed the way information is transmitted and stored. Renewable energy sources have become prevalent and affordable. Cooperation of the Information and Communication Technology industry and Renewable Energy industry makes it possible to create a next generation, decentralized power grid. In this context, the study seeks to identify the wider benefits to the local Japanese economy as a result of the development of a decentralised electricity market. Our general approach aims to integrate an economic analysis (monetary appraisal of costs and benefits to society) with externalities that are not quantifiable in monetary terms (e.g. social impact, environmental impact). The study also highlights opportunities and sets out recommendations for the citizens of the island and the local government. The simulation is the scientific basis for economic impact analysis. Various types of sources of energy have been taken into account: residential wind farm, residential wind turbine, solar farm, residential solar panels and private solar farms. Analysis of local geographic and economic conditions allowed creating a customized business model. Very often farmers on Awaji Island are using crop cycle. During each cycle, one part of the field is resting and replenishing nutrients. In the next year another part of the field is resting. Portable solar panels could be freely set up in this part of the field. At the end of the crop cycle, portable solar panels would be moved to the next resting part. Because of spacious area, for a single household 500 square meters of portable solar panels has been proposed and simulated. The devised simulation shows that the Rate of Return on Investment for solar panels, which are on the island, could reach up to 37.21%. Supposing that about 20% of households install solar panels they could produce 49.11% of the electric energy consumed by households on the island. The analysis shows that rest of the energy supply can be produced by currently existing one huge solar farm and two wind farms to meet 97.59% of demand on electricity for households on the island. Although there are more than 7,000 agricultural fields on the island, young people tend to avoid agricultural work and prefer to move from the island to big cities, live there in little mansions and work until late night. The business model proposed in this study could increase farmer’s monthly income by ¥200,000 - ¥300,000 (1,600 euro – 2,400 euro). Young people could work less and have a higher standard of living than in a city. Creation of a decentralized electricity market can unlock significant benefits in other industries (e.g. electric vehicles), providing a welcome boost to economic growth, jobs and quality of life.Keywords: digital twin, Matlab, model-based systems engineering, simulink, smart grid, systems engineering
Procedia PDF Downloads 1013047 Analyzing Factors Impacting COVID-19 Vaccination Rates
Authors: Dongseok Cho, Mitchell Driedger, Sera Han, Noman Khan, Mohammed Elmorsy, Mohamad El-Hajj
Abstract:
Since the approval of the COVID-19 vaccine in late 2020, vaccination rates have varied around the globe. Access to a vaccine supply, mandated vaccination policy, and vaccine hesitancy contribute to these rates. This study used COVID-19 vaccination data from Our World in Data and the Multilateral Leaders Task Force on COVID-19 to create two COVID-19 vaccination indices. The first index is the Vaccine Utilization Index (VUI), which measures how effectively each country has utilized its vaccine supply to doubly vaccinate its population. The second index is the Vaccination Acceleration Index (VAI), which evaluates how efficiently each country vaccinated its population within its first 150 days. Pearson correlations were created between these indices and country indicators obtained from the World Bank. The results of these correlations identify countries with stronger health indicators, such as lower mortality rates, lower age dependency ratios, and higher rates of immunization to other diseases, displaying higher VUI and VAI scores than countries with lesser values. VAI scores are also positively correlated to Governance and Economic indicators, such as regulatory quality, control of corruption, and GDP per capita. As represented by the VUI, proper utilization of the COVID-19 vaccine supply by country is observed in countries that display excellence in health practices. A country’s motivation to accelerate its vaccination rates within the first 150 days of vaccinating, as represented by the VAI, was largely a product of the governing body’s effectiveness and economic status, as well as overall excellence in health practises.Keywords: data mining, Pearson correlation, COVID-19, vaccination rates and hesitancy
Procedia PDF Downloads 1073046 Loan Repayment Prediction Using Machine Learning: Model Development, Django Web Integration and Cloud Deployment
Authors: Seun Mayowa Sunday
Abstract:
Loan prediction is one of the most significant and recognised fields of research in the banking, insurance, and the financial security industries. Some prediction systems on the market include the construction of static software. However, due to the fact that static software only operates with strictly regulated rules, they cannot aid customers beyond these limitations. Application of many machine learning (ML) techniques are required for loan prediction. Four separate machine learning models, random forest (RF), decision tree (DT), k-nearest neighbour (KNN), and logistic regression, are used to create the loan prediction model. Using the anaconda navigator and the required machine learning (ML) libraries, models are created and evaluated using the appropriate measuring metrics. From the finding, the random forest performs with the highest accuracy of 80.17% which was later implemented into the Django framework. For real-time testing, the web application is deployed on the Alibabacloud which is among the top 4 biggest cloud computing provider. Hence, to the best of our knowledge, this research will serve as the first academic paper which combines the model development and the Django framework, with the deployment into the Alibaba cloud computing application.Keywords: k-nearest neighbor, random forest, logistic regression, decision tree, django, cloud computing, alibaba cloud
Procedia PDF Downloads 1193045 Predictive Analytics of Student Performance Determinants
Authors: Mahtab Davari, Charles Edward Okon, Somayeh Aghanavesi
Abstract:
Every institute of learning is usually interested in the performance of enrolled students. The level of these performances determines the approach an institute of study may adopt in rendering academic services. The focus of this paper is to evaluate students' academic performance in given courses of study using machine learning methods. This study evaluated various supervised machine learning classification algorithms such as Logistic Regression (LR), Support Vector Machine, Random Forest, Decision Tree, K-Nearest Neighbors, Linear Discriminant Analysis, and Quadratic Discriminant Analysis, using selected features to predict study performance. The accuracy, precision, recall, and F1 score obtained from a 5-Fold Cross-Validation were used to determine the best classification algorithm to predict students’ performances. SVM (using a linear kernel), LDA, and LR were identified as the best-performing machine learning methods. Also, using the LR model, this study identified students' educational habits such as reading and paying attention in class as strong determinants for a student to have an above-average performance. Other important features include the academic history of the student and work. Demographic factors such as age, gender, high school graduation, etc., had no significant effect on a student's performance.Keywords: student performance, supervised machine learning, classification, cross-validation, prediction
Procedia PDF Downloads 1153044 A Unified Webcam Proctoring Solution on Edge
Authors: Saw Thiha, Jay Rajasekera
Abstract:
A boom in video conferencing generated millions of hours of video data daily to be analyzed. However, such enormous data pose certain scalability issues to be analyzed efficiently, let alone do it in real-time, as online conferences can involve hundreds of people and can last for hours. This paper proposes an efficient online proctoring solution that can analyze the online conferences real-time on edge devices such as Android, iOS, and desktops. Since the computation can be done upfront on the devices where online conferences take place, it can scale well without requiring intensive resources such as GPU servers and complex cloud infrastructure. According to the linear models, face orientation does indeed impact the perceived eye openness. Also, the proposed z score facial landmark standardization was proven to be functional in detecting face orientation and contributed to classifying eye blinks with single eyelid distance computation while achieving a better f1 score and accuracy than the Eye Aspect Ratio (EAR) threshold method. Last but not least, the authors implemented the solution natively in the MediaPipe framework and open-sourced it along with the reproducible experimental results on GitHub. The solution provides face orientation, eye blink, facial activity, and translation detections out of the box and is highly customizable and extensible.Keywords: android, desktop, edge computing, blink, face orientation, facial activity and translation, MediaPipe, open source, real-time, video conference, web, iOS, Z score facial landmark standardization
Procedia PDF Downloads 873043 Multimodal Employee Attendance Management System
Authors: Khaled Mohammed
Abstract:
This paper presents novel face recognition and identification approaches for the real-time attendance management problem in large companies/factories and government institutions. The proposed uses the Minimum Ratio (MR) approach for employee identification. Capturing the authentic face variability from a sequence of video frames has been considered for the recognition of faces and resulted in system robustness against the variability of facial features. Experimental results indicated an improvement in the performance of the proposed system compared to the Previous approaches at a rate between 2% to 5%. In addition, it decreased the time two times if compared with the Previous techniques, such as Extreme Learning Machine (ELM) & Multi-Scale Structural Similarity index (MS-SSIM). Finally, it achieved an accuracy of 99%.Keywords: attendance management system, face detection and recognition, live face recognition, minimum ratio
Procedia PDF Downloads 1473042 Random Forest Classification for Population Segmentation
Authors: Regina Chua
Abstract:
To reduce the costs of re-fielding a large survey, a Random Forest classifier was applied to measure the accuracy of classifying individuals into their assigned segments with the fewest possible questions. Given a long survey, one needed to determine the most predictive ten or fewer questions that would accurately assign new individuals to custom segments. Furthermore, the solution needed to be quick in its classification and usable in non-Python environments. In this paper, a supervised Random Forest classifier was modeled on a dataset with 7,000 individuals, 60 questions, and 254 features. The Random Forest consisted of an iterative collection of individual decision trees that result in a predicted segment with robust precision and recall scores compared to a single tree. A random 70-30 stratified sampling for training the algorithm was used, and accuracy trade-offs at different depths for each segment were identified. Ultimately, the Random Forest classifier performed at 87% accuracy at a depth of 10 with 20 instead of 254 features and 10 instead of 60 questions. With an acceptable accuracy in prioritizing feature selection, new tools were developed for non-Python environments: a worksheet with a formulaic version of the algorithm and an embedded function to predict the segment of an individual in real-time. Random Forest was determined to be an optimal classification model by its feature selection, performance, processing speed, and flexible application in other environments.Keywords: machine learning, supervised learning, data science, random forest, classification, prediction, predictive modeling
Procedia PDF Downloads 863041 Sentiment Analysis of Chinese Microblog Comments: Comparison between Support Vector Machine and Long Short-Term Memory
Authors: Xu Jiaqiao
Abstract:
Text sentiment analysis is an important branch of natural language processing. This technology is widely used in public opinion analysis and web surfing recommendations. At present, the mainstream sentiment analysis methods include three parts: sentiment analysis based on a sentiment dictionary, based on traditional machine learning, and based on deep learning. This paper mainly analyzes and compares the advantages and disadvantages of the SVM method of traditional machine learning and the Long Short-term Memory (LSTM) method of deep learning in the field of Chinese sentiment analysis, using Chinese comments on Sina Microblog as the data set. Firstly, this paper classifies and adds labels to the original comment dataset obtained by the web crawler, and then uses Jieba word segmentation to classify the original dataset and remove stop words. After that, this paper extracts text feature vectors and builds document word vectors to facilitate the training of the model. Finally, SVM and LSTM models are trained respectively. After accuracy calculation, it can be obtained that the accuracy of the LSTM model is 85.80%, while the accuracy of SVM is 91.07%. But at the same time, LSTM operation only needs 2.57 seconds, SVM model needs 6.06 seconds. Therefore, this paper concludes that: compared with the SVM model, the LSTM model is worse in accuracy but faster in processing speed.Keywords: sentiment analysis, support vector machine, long short-term memory, Chinese microblog comments
Procedia PDF Downloads 813040 Importance of New Policies of Process Management for Internet of Things Based on Forensic Investigation
Authors: Venkata Venugopal Rao Gudlur
Abstract:
The Proposed Policies referred to as “SOP”, on the Internet of Things (IoT) based Forensic Investigation into Process Management is the latest revolution to save time and quick solution for investigators. The forensic investigation process has been developed over many years from time to time it has been given the required information with no policies in investigation processes. This research reveals that the current IoT based forensic investigation into Process Management based is more connected to devices which is the latest revolution and policies. All future development in real-time information on gathering monitoring is evolved with smart sensor-based technologies connected directly to IoT. This paper present conceptual framework on process management. The smart devices are leading the way in terms of automated forensic models and frameworks established by different scholars. These models and frameworks were mostly focused on offering a roadmap for performing forensic operations with no policies in place. These initiatives would bring a tremendous benefit to process management and IoT forensic investigators proposing policies. The forensic investigation process may enhance more security and reduced data losses and vulnerabilities.Keywords: Internet of Things, Process Management, Forensic Investigation, M2M Framework
Procedia PDF Downloads 933039 Characterization of Internet Exchange Points by Using Quantitative Data
Authors: Yamba Dabone, Tounwendyam Frédéric Ouedraogo, Pengwendé Justin Kouraogo, Oumarou Sie
Abstract:
Reliable data transport over the Internet is one of the goals of researchers in the field of computer science. Data such as videos and audio files are becoming increasingly large. As a result, transporting them over the Internet is becoming difficult. Therefore, it has been important to establish a method to locally interconnect autonomous systems (AS) with each other to facilitate traffic exchange. It is in this context that Internet Exchange Points (IXPs) are set up to facilitate local and even regional traffic. They are now the lifeblood of the Internet. Therefore, it is important to think about the factors that can characterize IXPs. However, other more quantifiable characteristics can help determine the quality of an IXP. In addition, these characteristics may allow ISPs to have a clearer view of the exchange node and may also convince other networks to connect to an IXP. To that end, we define five new IXP characteristics: the attraction rate (τₐₜₜᵣ); and the peering rate (τₚₑₑᵣ); the target rate of an IXP (Objₐₜₜ); the number of IXP links (Nₗᵢₙₖ); the resistance rate τₑ𝒻𝒻 and the attraction failure rate (τ𝒻).Keywords: characteristic, autonomous system, internet service provider, internet exchange point, rate
Procedia PDF Downloads 863038 A Review: Detection and Classification Defects on Banana and Apples by Computer Vision
Authors: Zahow Muoftah
Abstract:
Traditional manual visual grading of fruits has been one of the agricultural industry’s major challenges due to its laborious nature as well as inconsistency in the inspection and classification process. The main requirements for computer vision and visual processing are some effective techniques for identifying defects and estimating defect areas. Automated defect detection using computer vision and machine learning has emerged as a promising area of research with a high and direct impact on the visual inspection domain. Grading, sorting, and disease detection are important factors in determining the quality of fruits after harvest. Many studies have used computer vision to evaluate the quality level of fruits during post-harvest. Many studies have used computer vision to evaluate the quality level of fruits during post-harvest. Many studies have been conducted to identify diseases and pests that affect the fruits of agricultural crops. However, most previous studies concentrated solely on the diagnosis of a lesion or disease. This study focused on a comprehensive study to identify pests and diseases of apple and banana fruits using detection and classification defects on Banana and Apples by Computer Vision. As a result, the current article includes research from these domains as well. Finally, various pattern recognition techniques for detecting apple and banana defects are discussed.Keywords: computer vision, banana, apple, detection, classification
Procedia PDF Downloads 903037 Enhancer: An Effective Transformer Architecture for Single Image Super Resolution
Authors: Pitigalage Chamath Chandira Peiris
Abstract:
A widely researched domain in the field of image processing in recent times has been single image super-resolution, which tries to restore a high-resolution image from a single low-resolution image. Many more single image super-resolution efforts have been completed utilizing equally traditional and deep learning methodologies, as well as a variety of other methodologies. Deep learning-based super-resolution methods, in particular, have received significant interest. As of now, the most advanced image restoration approaches are based on convolutional neural networks; nevertheless, only a few efforts have been performed using Transformers, which have demonstrated excellent performance on high-level vision tasks. The effectiveness of CNN-based algorithms in image super-resolution has been impressive. However, these methods cannot completely capture the non-local features of the data. Enhancer is a simple yet powerful Transformer-based approach for enhancing the resolution of images. A method for single image super-resolution was developed in this study, which utilized an efficient and effective transformer design. This proposed architecture makes use of a locally enhanced window transformer block to alleviate the enormous computational load associated with non-overlapping window-based self-attention. Additionally, it incorporates depth-wise convolution in the feed-forward network to enhance its ability to capture local context. This study is assessed by comparing the results obtained for popular datasets to those obtained by other techniques in the domain.Keywords: single image super resolution, computer vision, vision transformers, image restoration
Procedia PDF Downloads 963036 Virtual Reality as a Tool in Modern Education
Authors: Łukasz Bis
Abstract:
The author is going to discuss virtual reality and its importance for new didactic methods. It has been known for years that experience-based education gives much better results in terms of long-term memory than theoretical study. However, practice is expensive - virtual reality allows the use of an empirical approach to learning, with minimized production costs. The author defines what makes a given VR experience appropriate (adequate) for the didactic and cognitive process. The article is a kind of a list of guidelines and their importance for the VR experience under development.Keywords: virtual reality, education, universal design, guideline
Procedia PDF Downloads 973035 Proposal of a Virtual Reality Dynamism Augmentation Method for Sports Spectating
Authors: Hertzog Clara, Sakurai Sho, Hirota Koichi, Nojima Takuya
Abstract:
It is common to see graphics appearing on television while watching a sports game to provide information, but it is less common to see graphics specifically aiming to boost spectators’ dynamism perception. It is even less common to see such graphics designed especially for virtual reality (VR). However, it appears that even with simple dynamic graphics, it would be possible to improve VR sports spectators’ experience. So, in this research, we explain how graphics can be used in VR to improve the dynamism of a broadcasted sports game and we provide a simple example. This example consists in a white halo displayed around the video and blinking according to the game speed. We hope to increase people’s awareness about VR sports spectating and the possibilities this display offers through dynamic graphics.Keywords: broadcasting, graphics, sports spectating, virtual reality
Procedia PDF Downloads 823034 Efficient Heuristic Algorithm to Speed Up Graphcut in Gpu for Image Stitching
Authors: Tai Nguyen, Minh Bui, Huong Ninh, Tu Nguyen, Hai Tran
Abstract:
GraphCut algorithm has been widely utilized to solve various types of computer vision problems. Its expensive computational cost encouraged many researchers to improve the speed of the algorithm. Recent works proposed schemes that work on parallel computing platforms such as CUDA. However, the problem of low convergence speed prevents the usage of GraphCut for real time applications. In this paper, we propose global suppression heuristic to boost the conver-gence process of the algorithm. A parallel implementation of GraphCut algorithm on CUDA designed for the image stitching problem is introduced. Our method achieves up to 3× time boost on the graph of size 80 × 480 compared to the best sequential GraphCut algorithm while achieving satisfactory stitched images, suitable for panorama applications. Our source code will be soon available for further research.Keywords: CUDA, graph cut, image stitching, texture synthesis, maxflow/mincut algorithm
Procedia PDF Downloads 1203033 Network Word Discovery Framework Based on Sentence Semantic Vector Similarity
Authors: Ganfeng Yu, Yuefeng Ma, Shanliang Yang
Abstract:
The word discovery is a key problem in text information retrieval technology. Methods in new word discovery tend to be closely related to words because they generally obtain new word results by analyzing words. With the popularity of social networks, individual netizens and online self-media have generated various network texts for the convenience of online life, including network words that are far from standard Chinese expression. How detect network words is one of the important goals in the field of text information retrieval today. In this paper, we integrate the word embedding model and clustering methods to propose a network word discovery framework based on sentence semantic similarity (S³-NWD) to detect network words effectively from the corpus. This framework constructs sentence semantic vectors through a distributed representation model, uses the similarity of sentence semantic vectors to determine the semantic relationship between sentences, and finally realizes network word discovery by the meaning of semantic replacement between sentences. The experiment verifies that the framework not only completes the rapid discovery of network words but also realizes the standard word meaning of the discovery of network words, which reflects the effectiveness of our work.Keywords: text information retrieval, natural language processing, new word discovery, information extraction
Procedia PDF Downloads 863032 Neural Rendering Applied to Confocal Microscopy Images
Authors: Daniel Li
Abstract:
We present a novel application of neural rendering methods to confocal microscopy. Neural rendering and implicit neural representations have developed at a remarkable pace, and are prevalent in modern 3D computer vision literature. However, they have not yet been applied to optical microscopy, an important imaging field where 3D volume information may be heavily sought after. In this paper, we employ neural rendering on confocal microscopy focus stack data and share the results. We highlight the benefits and potential of adding neural rendering to the toolkit of microscopy image processing techniques.Keywords: neural rendering, implicit neural representations, confocal microscopy, medical image processing
Procedia PDF Downloads 6493031 Effective Stacking of Deep Neural Models for Automated Object Recognition in Retail Stores
Authors: Ankit Sinha, Soham Banerjee, Pratik Chattopadhyay
Abstract:
Automated product recognition in retail stores is an important real-world application in the domain of Computer Vision and Pattern Recognition. In this paper, we consider the problem of automatically identifying the classes of the products placed on racks in retail stores from an image of the rack and information about the query/product images. We improve upon the existing approaches in terms of effectiveness and memory requirement by developing a two-stage object detection and recognition pipeline comprising of a Faster-RCNN-based object localizer that detects the object regions in the rack image and a ResNet-18-based image encoder that classifies the detected regions into the appropriate classes. Each of the models is fine-tuned using appropriate data sets for better prediction and data augmentation is performed on each query image to prepare an extensive gallery set for fine-tuning the ResNet-18-based product recognition model. This encoder is trained using a triplet loss function following the strategy of online-hard-negative-mining for improved prediction. The proposed models are lightweight and can be connected in an end-to-end manner during deployment to automatically identify each product object placed in a rack image. Extensive experiments using Grozi-32k and GP-180 data sets verify the effectiveness of the proposed model.Keywords: retail stores, faster-RCNN, object localization, ResNet-18, triplet loss, data augmentation, product recognition
Procedia PDF Downloads 1403030 Advancement of Computer Science Research in Nigeria: A Bibliometric Analysis of the Past Three Decades
Authors: Temidayo O. Omotehinwa, David O. Oyewola, Friday J. Agbo
Abstract:
This study aims to gather a proper perspective of the development landscape of Computer Science research in Nigeria. Therefore, a bibliometric analysis of 4,333 bibliographic records of Computer Science research in Nigeria in the last 31 years (1991-2021) was carried out. The bibliographic data were extracted from the Scopus database and analyzed using VOSviewer and the bibliometrix R package through the biblioshiny web interface. The findings of this study revealed that Computer Science research in Nigeria has a growth rate of 24.19%. The most developed and well-studied research areas in the Computer Science field in Nigeria are machine learning, data mining, and deep learning. The social structure analysis result revealed that there is a need for improved international collaborations. Sparsely established collaborations are largely influenced by geographic proximity. The funding analysis result showed that Computer Science research in Nigeria is under-funded. The findings of this study will be useful for researchers conducting Computer Science related research. Experts can gain insights into how to develop a strategic framework that will advance the field in a more impactful manner. Government agencies and policymakers can also utilize the outcome of this research to develop strategies for improved funding for Computer Science research.Keywords: bibliometric analysis, biblioshiny, computer science, Nigeria, science mapping
Procedia PDF Downloads 973029 Time-Series Load Data Analysis for User Power Profiling
Authors: Mahdi Daghmhehci Firoozjaei, Minchang Kim, Dima Alhadidi
Abstract:
In this paper, we present a power profiling model for smart grid consumers based on real time load data acquired smart meters. It profiles consumers’ power consumption behaviour using the dynamic time warping (DTW) clustering algorithm. Due to the invariability of signal warping of this algorithm, time-disordered load data can be profiled and consumption features be extracted. Two load types are defined and the related load patterns are extracted for classifying consumption behaviour by DTW. The classification methodology is discussed in detail. To evaluate the performance of the method, we analyze the time-series load data measured by a smart meter in a real case. The results verify the effectiveness of the proposed profiling method with 90.91% true positive rate for load type clustering in the best case.Keywords: power profiling, user privacy, dynamic time warping, smart grid
Procedia PDF Downloads 1373028 Vehicle Speed Estimation Using Image Processing
Authors: Prodipta Bhowmik, Poulami Saha, Preety Mehra, Yogesh Soni, Triloki Nath Jha
Abstract:
In India, the smart city concept is growing day by day. So, for smart city development, a better traffic management and monitoring system is a very important requirement. Nowadays, road accidents increase due to more vehicles on the road. Reckless driving is mainly responsible for a huge number of accidents. So, an efficient traffic management system is required for all kinds of roads to control the traffic speed. The speed limit varies from road to road basis. Previously, there was a radar system but due to high cost and less precision, the radar system is unable to become favorable in a traffic management system. Traffic management system faces different types of problems every day and it has become a researchable topic on how to solve this problem. This paper proposed a computer vision and machine learning-based automated system for multiple vehicle detection, tracking, and speed estimation of vehicles using image processing. Detection of vehicles and estimating their speed from a real-time video is tough work to do. The objective of this paper is to detect vehicles and estimate their speed as accurately as possible. So for this, a real-time video is first captured, then the frames are extracted from that video, then from that frames, the vehicles are detected, and thereafter, the tracking of vehicles starts, and finally, the speed of the moving vehicles is estimated. The goal of this method is to develop a cost-friendly system that can able to detect multiple types of vehicles at the same time.Keywords: OpenCV, Haar Cascade classifier, DLIB, YOLOV3, centroid tracker, vehicle detection, vehicle tracking, vehicle speed estimation, computer vision
Procedia PDF Downloads 713027 Developing a Video Game (Historia’s Nightmare) and Finding Out if We Can Use It to Raise Social Awareness and Improve Learning
Authors: Hasibul Kabir, Samin Shahriar Tokey, Md. Tofazzal Hossain
Abstract:
One of the most necessary things in the present time is raising social awareness about global warming and climate change among the people. Though many types of mediums and techniques have been used to teach people about this global phenomenon, there are still more effective ways to reach people with useful information about global warming. As many traditional methods to teach people about global warming and climate change did not work well, video games were overdue. To learn how effective a video game can be in this regard, we developed a Video game, "Historia's Nightmare," that teaches people about Global warming and climate change. The game was designed to entertain people and give them an idea about the reasons and consequences of global warming and climate change while not being like traditional educational games. The game threw a mini quiz consisting of two MCQs based on the information shown in the game, where a gamer had to pass the quiz to reach the next level. We published the game on different platforms to let all types of people play and complete our experiment effectively. The game continuously communicated with our server to send data about gamers' performance. We observed the data, including the participants' performance, time spent, quiz score, and the in-game feedback on a regular basis, and finally came to a verdict. In our experiment, we have found that most participants positively accepted the game and learned something new. The participants who spent more on our game performed better in both quiz and the game. Our experiment's result demonstrates that video games can be a great way to teach people something, particularly to raise social awareness about global warming and climate change. It also demonstrates that the game can be a significant element in education and learning improvement.Keywords: video game, global warming, social awareness, climate change, education, feedback
Procedia PDF Downloads 1233026 Genetic Testing and Research in South Africa: The Sharing of Data Across Borders
Authors: Amy Gooden, Meshandren Naidoo
Abstract:
Genetic research is not confined to a particular jurisdiction. Using direct-to-consumer genetic testing (DTC-GT) as an example, this research assesses the status of data sharing into and out of South Africa (SA). While SA laws cover the sending of genetic data out of SA, prohibiting such transfer unless a legal ground exists, the position where genetic data comes into the country depends on the laws of the country from where it is sent – making the legal position less clear.Keywords: cross-border, data, genetic testing, law, regulation, research, sharing, South Africa
Procedia PDF Downloads 1493025 Masked Candlestick Model: A Pre-Trained Model for Trading Prediction
Authors: Ling Qi, Matloob Khushi, Josiah Poon
Abstract:
This paper introduces a pre-trained Masked Candlestick Model (MCM) for trading time-series data. The pre-trained model is based on three core designs. First, we convert trading price data at each data point as a set of normalized elements and produce embeddings of each element. Second, we generate a masked sequence of such embedded elements as inputs for self-supervised learning. Third, we use the encoder mechanism from the transformer to train the inputs. The masked model learns the contextual relations among the sequence of embedded elements, which can aid downstream classification tasks. To evaluate the performance of the pre-trained model, we fine-tune MCM for three different downstream classification tasks to predict future price trends. The fine-tuned models achieved better accuracy rates for all three tasks than the baseline models. To better analyze the effectiveness of MCM, we test the same architecture for three currency pairs, namely EUR/GBP, AUD/USD, and EUR/JPY. The experimentation results demonstrate MCM’s effectiveness on all three currency pairs and indicate the MCM’s capability for signal extraction from trading data.Keywords: masked language model, transformer, time series prediction, trading prediction, embedding, transfer learning, self-supervised learning
Procedia PDF Downloads 1143024 Evolutionary Advantages of Loneliness with an Agent-Based Model
Authors: David Gottlieb, Jason Yoder
Abstract:
The feeling of loneliness is not uncommon in modern society, and yet, there is a fundamental lack of understanding in its origins and purpose in nature. One interpretation of loneliness is that it is a subjective experience that punishes a lack of social behavior, and thus its emergence in human evolution is seemingly tied to the survival of early human tribes. Still, a common counterintuitive response to loneliness is a state of hypervigilance, resulting in social withdrawal, which may appear maladaptive to modern society. So far, no computational model of loneliness’ effect during evolution yet exists; however, agent-based models (ABM) can be used to investigate social behavior, and applying evolution to agents’ behaviors can demonstrate selective advantages for particular behaviors. We propose an ABM where each agent contains four social behaviors, and one goal-seeking behavior, letting evolution select the best behavioral patterns for resource allocation. In our paper, we use an algorithm similar to the boid model to guide the behavior of agents, but expand the set of rules that govern their behavior. While we use cohesion, separation, and alignment for simple social movement, our expanded model adds goal-oriented behavior, which is inspired by particle swarm optimization, such that agents move relative to their personal best position. Since agents are given the ability to form connections by interacting with each other, our final behavior guides agent movement toward its social connections. Finally, we introduce a mechanism to represent a state of loneliness, which engages when an agent's perceived social involvement does not meet its expected social involvement. This enables us to investigate a minimal model of loneliness, and using evolution we attempt to elucidate its value in human survival. Agents are placed in an environment in which they must acquire resources, as their fitness is based on the total resource collected. With these rules in place, we are able to run evolution under various conditions, including resource-rich environments, and when disease is present. Our simulations indicate that there is strong selection pressure for social behavior under circumstances where there is a clear discrepancy between initial resource locations, and against social behavior when disease is present, mirroring hypervigilance. This not only provides an explanation for the emergence of loneliness, but also reflects the diversity of response to loneliness in the real world. In addition, there is evidence of a richness of social behavior when loneliness was present. By introducing just two resource locations, we observed a divergence in social motivation after agents became lonely, where one agent learned to move to the other, who was in a better resource position. The results and ongoing work from this project show that it is possible to glean insight into the evolutionary advantages of even simple mechanisms of loneliness. The model we developed has produced unexpected results and has led to more questions, such as the impact loneliness would have at a larger scale, or the effect of creating a set of rules governing interaction beyond adjacency.Keywords: agent-based, behavior, evolution, loneliness, social
Procedia PDF Downloads 91