Search results for: computational accuracy

Authors: Hanik Park, Kyungsook Jeon, Suchul Shin, Youngchul Park

Abstract:

The air vent head prevents the inflow of seawater into the cargo holds when it is used for the ballast tank on heavy weather. In this study, the flow characteristics and the grid size were created by the application of Computational Fluid Dynamics by taking into the consideration of comparison of test results. Then, the accuracy of the analysis was verified by comparing with experimental results. Based on this analysis, accurate turbulence model and grid size can be selected. Thus, the design characteristic of air vent head for bulk carrier contributes the reliability based on the research results.

Keywords: bulk carrier, FEM, SST, vent

Procedia PDF Downloads 518

Authors: Aymen Laadhari, Gábor Székely

Abstract:

In this work, the hemodynamics in the sinuses of Valsalva after Transcatheter Aortic Valve Implantation is numerically examined. We focus on the physical results in the two-dimensional case. We use a finite element methodology based on a Lagrange multiplier technique that enables to couple the dynamics of blood flow and the leaflets’ movement. A massively parallel implementation of a monolithic and fully implicit solver allows more accuracy and significant computational savings. The elastic properties of the aortic valve are disregarded, and the numerical computations are performed under physiologically correct pressure loads. Computational results depict that blood flow may be subject to stagnation in the lower domain of the sinuses of Valsalva after Transcatheter Aortic Valve Implantation.

Keywords: hemodynamics, simulations, stagnation, valve

Procedia PDF Downloads 291

Authors: Stewart A. Keir, Faik A. Hamad

Abstract:

This study aims to use both analytical and experimental methods of analysis to examine the accuracy of Computational Fluid Dynamics (CFD) models that can then be used for more complex analyses, accurately representing more elaborate flow phenomena such as internal shockwaves and boundary layers. The geometry used in the analytical study and CFD model is taken from the experimental rig. The analytical study is undertaken using isentropic and adiabatic relationships and the output of the analytical study, the 'shockwave location tool', is created. The results from the analytical study are then used to optimize the redesign an experimental rig for more favorable placement of pressure taps and gain a much better representation of the shockwaves occurring in the divergent section of the nozzle. The CFD model is then optimized through the selection of different parameters, e.g. turbulence models (Spalart-Almaras, Realizable k-epsilon & Standard k-omega) in order to develop an accurate, robust model. The results from the CFD model can then be directly compared to experimental and analytical results in order to gauge the accuracy of each method of analysis. The CFD model will be used to visualize the variation of various parameters such as velocity/Mach number, pressure and turbulence across the shock. The CFD results will be used to investigate the interaction between the shock wave and the boundary layer. The validated model can then be used to modify the nozzle designs which may offer better performance and ease of manufacture and may present feasible improvements to existing high-speed flow applications.

Keywords: CFD, nozzle, fluent, gas dynamics, shock-wave

Procedia PDF Downloads 233

Authors: Albert Acheampong, Tamer Elshandidy

Abstract:

We investigate the extent to which sustainability disclosure from the narrative section of European banks’ annual reports improves analyst forecast accuracy. We capture sustainability disclosure using a machine learning approach and use forecast error to proxy analyst forecast accuracy. Our results suggest that sustainability disclosure significantly improves analyst forecast accuracy by reducing the forecast error. In a further analysis, we also find that the induction of Directive 2014/95/European Union (EU) is associated with increased disclosure content, which then reduces forecast error. Collectively, our results suggest that sustainability disclosure improves forecast accuracy, and the induction of the new EU directive strengthens this improvement. These results hold after several further and robustness analyses. Our findings have implications for market participants and policymakers.

Keywords: sustainability disclosure, machine learning, analyst forecast accuracy, forecast error, European banks, EU directive

Procedia PDF Downloads 76

Authors: Rajvir Kaur, Jeewani Anupama Ginige

Abstract:

With recent trends in Big Data and advancements in Information and Communication Technologies, the healthcare industry is at the stage of its transition from clinician oriented to technology oriented. Many people around the world die of cancer because the diagnosis of disease was not done at an early stage. Nowadays, the computational methods in the form of Machine Learning (ML) are used to develop automated decision support systems that can diagnose cancer with high confidence in a timely manner. This paper aims to carry out the comparative evaluation of a selected set of ML classifiers on two existing datasets: breast cancer and cervical cancer. The ML classifiers compared in this study are Decision Tree (DT), Support Vector Machine (SVM), k-Nearest Neighbor (k-NN), Logistic Regression, Ensemble (Bagged Tree) and Artificial Neural Networks (ANN). The evaluation is carried out based on standard evaluation metrics Precision (P), Recall (R), F1-score and Accuracy. The experimental results based on the evaluation metrics show that ANN showed the highest-level accuracy (99.4%) when tested with breast cancer dataset. On the other hand, when these ML classifiers are tested with the cervical cancer dataset, Ensemble (Bagged Tree) technique gave better accuracy (93.1%) in comparison to other classifiers.

Keywords: artificial neural networks, breast cancer, classifiers, cervical cancer, f-score, machine learning, precision, recall

Procedia PDF Downloads 277

Authors: Abdullah Alhomidan

Abstract:

This study is conducted to identify the main factors affecting accuracy of pretender cost estimate in building construction projects in Saudi Arabia from owners’ perspective. 44 factors affecting pretender cost estimate were identified through literature review and discussion with some construction experts. The results show that the top important factors affecting pretender cost estimate accuracy are: level of competitors in the tendering, material price changes, communications with suppliers, communications with client, and estimating method used.

Keywords: cost estimate, accuracy, pretender, estimating, bid estimate

Procedia PDF Downloads 556

Authors: Thomas Bryan, Veton Kepuska, Ivica Kostanic

Abstract:

Speaker recognition is performed in high Additive White Gaussian Noise (AWGN) environments using principals of Computational Auditory Scene Analysis (CASA). CASA methods often classify sounds from images in the time-frequency (T-F) plane using spectrograms or cochleargrams as the image. In this paper atomic decomposition implemented by matching pursuit performs a transform from time series speech signals to the T-F plane. The atomic decomposition creates a sparsely populated T-F vector in “weight space” where each populated T-F position contains an amplitude weight. The weight space vector along with the atomic dictionary represents a denoised, compressed version of the original signal. The arraignment or of the atomic indices in the T-F vector are used for classification. Unsupervised feature learning implemented by a sparse autoencoder learns a single dictionary of basis features from a collection of envelope samples from all speakers. The approach is demonstrated using pairs of speakers from the TIMIT data set. Pairs of speakers are selected randomly from a single district. Each speak has 10 sentences. Two are used for training and 8 for testing. Atomic index probabilities are created for each training sentence and also for each test sentence. Classification is performed by finding the lowest Euclidean distance between then probabilities from the training sentences and the test sentences. Training is done at a 30dB Signal-to-Noise Ratio (SNR). Testing is performed at SNR’s of 0 dB, 5 dB, 10 dB and 30dB. The algorithm has a baseline classification accuracy of ~93% averaged over 10 pairs of speakers from the TIMIT data set. The baseline accuracy is attributable to short sequences of training and test data as well as the overall simplicity of the classification algorithm. The accuracy is not affected by AWGN and produces ~93% accuracy at 0dB SNR.

Keywords: time-frequency plane, atomic decomposition, envelope sampling, Gabor atoms, matching pursuit, sparse dictionary learning, sparse autoencoder

Procedia PDF Downloads 289

Authors: Mintu Philip, Renumol V. G.

Abstract:

An introductory programming course is a major challenge faced in Computing Education. Many of the introductory programming courses fail because student concentrate mainly on writing programs using a programming language rather than involving in problem solving. Computational thinking is a general approach to solve problems. This paper proposes a new preliminary course that aims to develop computational thinking skills in students, which may help them to become good programmers. The proposed course is designed based on the four basic components of computational thinking - abstract thinking, logical thinking, modeling thinking and constructive thinking. In this course, students are engaged in hands-on problem solving activities using a new problem solving model proposed in this paper.

Keywords: computational thinking, computing education, abstraction, constructive thinking, modelling thinking

Procedia PDF Downloads 456

Authors: Gebreegziabher Hailu

Abstract:

This paper presents the development of residual power series methods aimed at efficiently solving stiff differential equations, which pose significant challenges in numerical analysis due to their rapid changes in solution behavior. The RPSM is a numerical approach that generates polynomial-based approximate solutions without the need for linearization, discretization, or perturbation techniques, making it straightforward to implement and less prone to computational errors. We introduce an approach that utilizes power series expansions combined with residual minimization techniques to enhance convergence and stability. By analyzing the theoretical foundations of stiffness, we delve into the formulation of the residual power series method, detailing how it effectively captures the dynamics of stiff systems while maintaining computational efficiency. Numerical experiments demonstrate the method's superiority in terms of accuracy and computational cost when compared to traditional methods like implicit Runge-Kutta or multistep techniques. We also explore adaptive strategies within our framework to automatically adjust parameters based on the stiffness characteristics of the problem at hand. Ultimately, our findings contribute to the broader toolkit for tackling stiff differential equations, offering a robust alternative that promises to streamline computational workflows in various applied mathematics and engineering contexts.

Keywords: residual power series methods, stiff differential equoations, numerical approach, Runge Kutta methods

Procedia PDF Downloads 23

Authors: Qian Yi Ooi

Abstract:

At present, airfoil parameters are still designed and optimized according to the scale of conventional aircraft, and there are still some slight deviations in terms of scale differences. However, insufficient parameters or poor surface mesh quality is likely to occur if these small deviations are embedded in a future civil aircraft with a size that is quite different from conventional aircraft, such as a blended-wing-body (BWB) aircraft with future potential, resulting in large deviations in geometric similarity in computational fluid dynamics (CFD) simulations. To avoid this situation, the study on the CFD calculation on the geometric similarity of airfoil parameters and the quality of the surface mesh is conducted to obtain the ability of different parameterization methods applied on different airfoil scales. The research objects are three airfoil scales, including the wing root and wingtip of conventional civil aircraft and the wing root of the giant hybrid wing, used by three parameterization methods to compare the calculation differences between different sizes of airfoils. In this study, the constants including NACA 0012, a Reynolds number of 10 million, an angle of attack of zero, a C-grid for meshing, and the k-epsilon (k-ε) turbulence model are used. The experimental variables include three airfoil parameterization methods: point cloud method, B-spline curve method, and class function/shape function transformation (CST) method. The airfoil dimensions are set to 3.98 meters, 17.67 meters, and 48 meters, respectively. In addition, this study also uses different numbers of edge meshing and the same bias factor in the CFD simulation. Studies have shown that with the change of airfoil scales, different parameterization methods, the number of control points, and the meshing number of divisions should be used to improve the accuracy of the aerodynamic performance of the wing. When the airfoil ratio increases, the most basic point cloud parameterization method will require more and larger data to support the accuracy of the airfoil’s aerodynamic performance, which will face the severe test of insufficient computer capacity. On the other hand, when using the B-spline curve method, average number of control points and meshing number of divisions should be set appropriately to obtain higher accuracy; however, the quantitative balance cannot be directly defined, but the decisions should be made repeatedly by adding and subtracting. Lastly, when using the CST method, it is found that limited control points are enough to accurately parameterize the larger-sized wing; a higher degree of accuracy and stability can be obtained by using a lower-performance computer.

Keywords: airfoil, computational fluid dynamics, geometric similarity, surface mesh quality

Procedia PDF Downloads 222

Authors: Semeh Ben Salem, Sami Naouali, Moetez Sallami

Abstract:

The increasing amount of collected data has limited the performance of the current analyzing algorithms. Thus, developing new cost-effective algorithms in terms of complexity, scalability, and accuracy raised significant interests. In this paper, a modified effective k-means based algorithm is developed and experimented. The new algorithm aims to reduce the computational load without significantly affecting the quality of the clusterings. The algorithm uses the City Block distance and a new stop criterion to guarantee the convergence. Conducted experiments on a real data set show its high performance when compared with the original k-means version.

Keywords: pattern recognition, global terrorism database, Manhattan distance, k-means clustering, terrorism data analysis

Procedia PDF Downloads 386

Authors: Dimitri Golenko-Ginzburg

Abstract:

A CPM network project with deterministic activity durations, in which activities require homogenous resources with fixed capacities, is considered. The problem is to determine the optimal schedule of starting times for all network activities within their maximal allowable limits (in order not to exceed the network's critical time) to minimize the maximum required resources for the project at any point in time. In case when a non-critical activity may start only at discrete moments with the pregiven time span, the problem becomes NP-complete and an optimal solution may be obtained via a look-over algorithm. For the case when a look-over requires much computational time an approximate algorithm is suggested. The algorithm's performance ratio, i.e., the relative accuracy error, is determined. Experimentation has been undertaken to verify the suggested algorithm.

Keywords: resource smoothing problem, CPM network, lookover algorithm, lexicographical order, approximate algorithm, accuracy estimate

Procedia PDF Downloads 302

Authors: M. Pienaar, J. W. Odendaal, J. C. Smit, J. Joubert

Abstract:

Simulations are commonly used to predict the bistatic radar cross section (RCS) of military targets since characterization measurements can be expensive and time consuming. It is thus important to accurately predict the bistatic RCS of targets. Computational electromagnetic (CEM) methods can be used for bistatic RCS prediction. CEM methods are divided into full-wave and asymptotic methods. Full-wave methods are numerical approximations to the exact solution of Maxwell’s equations. These methods are very accurate but are computationally very intensive and time consuming. Asymptotic techniques make simplifying assumptions in solving Maxwell's equations and are thus less accurate but require less computational resources and time. Asymptotic techniques can thus be very valuable for the prediction of bistatic RCS of electrically large targets, due to the decreased computational requirements. This study extends previous work by validating the accuracy of asymptotic techniques to predict bistatic RCS through comparison with full-wave simulations as well as measurements. Validation is done with canonical structures as well as complex realistic aircraft models instead of only looking at a complex slicy structure. The slicy structure is a combination of canonical structures, including cylinders, corner reflectors and cubes. Validation is done over large bistatic angles and at different polarizations. Bistatic RCS measurements were conducted in a compact range, at the University of Pretoria, South Africa. The measurements were performed at different polarizations from 2 GHz to 6 GHz. Fixed bistatic angles of β = 30.8°, 45° and 90° were used. The measurements were calibrated with an active calibration target. The EM simulation tool FEKO was used to generate simulated results. The full-wave multi-level fast multipole method (MLFMM) simulated results together with the measured data were used as reference for validation. The accuracy of physical optics (PO) and geometrical optics (GO) was investigated. Differences relating to amplitude, lobing structure and null positions were observed between the asymptotic, full-wave and measured data. PO and GO were more accurate at angles close to the specular scattering directions and the accuracy seemed to decrease as the bistatic angle increased. At large bistatic angles PO did not perform well due to the shadow regions not being treated appropriately. PO also did not perform well for canonical structures where multi-bounce was the main scattering mechanism. PO and GO do not account for diffraction but these inaccuracies tended to decrease as the electrical size of objects increased. It was evident that both asymptotic techniques do not properly account for bistatic structural shadowing. Specular scattering was calculated accurately even if targets did not meet the electrically large criteria. It was evident that the bistatic RCS prediction performance of PO and GO depends on incident angle, frequency, target shape and observation angle. The improved computational efficiency of the asymptotic solvers yields a major advantage over full-wave solvers and measurements; however, there is still much room for improvement of the accuracy of these asymptotic techniques.

Keywords: asymptotic techniques, bistatic RCS, geometrical optics, physical optics

Procedia PDF Downloads 258

Authors: Paramjit Kaur, Harish Kundra

Abstract:

In the recent years, swarm intelligence based firefly algorithm has become a great focus for the researchers to solve the real time optimization problems. Here, firefly algorithm is used for the application of satellite image classification. For experimentation, Alwar area is considered to multiple land features like vegetation, barren, hilly, residential and water surface. Alwar dataset is considered with seven band satellite images. Firefly Algorithm is based on the attraction of less bright fireflies towards more brightener one. For the evaluation of proposed concept accuracy assessment parameters are calculated using error matrix. With the help of Error matrix, parameters of Kappa Coefficient, Overall Accuracy and feature wise accuracy parameters of user’s accuracy & producer’s accuracy can be calculated. Overall results are compared with BBO, PSO, Hybrid FPAB/BBO, Hybrid ACO/SOFM and Hybrid ACO/BBO based on the kappa coefficient and overall accuracy parameters.

Keywords: image classification, firefly algorithm, satellite image classification, terrain classification

Procedia PDF Downloads 401

Authors: Jack R. McKenzie, Peter A. Appleby, Thomas House, Neil Walton

Abstract:

Cold-start is a notoriously difficult problem which can occur in recommendation systems, and arises when there is insufficient information to draw inferences for users or items. To address this challenge, a contextual bandit algorithm – the Fast Approximate Bayesian Contextual Cold Start Learning algorithm (FAB-COST) – is proposed, which is designed to provide improved accuracy compared to the traditionally used Laplace approximation in the logistic contextual bandit, while controlling both algorithmic complexity and computational cost. To this end, FAB-COST uses a combination of two moment projection variational methods: Expectation Propagation (EP), which performs well at the cold start, but becomes slow as the amount of data increases; and Assumed Density Filtering (ADF), which has slower growth of computational cost with data size but requires more data to obtain an acceptable level of accuracy. By switching from EP to ADF when the dataset becomes large, it is able to exploit their complementary strengths. The empirical justification for FAB-COST is presented, and systematically compared to other approaches on simulated data. In a benchmark against the Laplace approximation on real data consisting of over 670, 000 impressions from autotrader.co.uk, FAB-COST demonstrates at one point increase of over 16% in user clicks. On the basis of these results, it is argued that FAB-COST is likely to be an attractive approach to cold-start recommendation systems in a variety of contexts.

Keywords: cold-start learning, expectation propagation, multi-armed bandits, Thompson Sampling, variational inference

Procedia PDF Downloads 108

Authors: Wei Sun, Yan Dong

Abstract:

There is an increasing interest in introducing computational thinking at an early age. Computational thinking, like mathematical thinking, engineering thinking, and scientific thinking, is a kind of analytical thinking. Learning computational thinking skills is not only to improve technological literacy, but also allows learners to equip with practicable skills such as problem-solving skills. As people realize the importance of computational thinking, the field of educational technology faces a problem: how to choose appropriate tools and activities to help students develop computational thinking skills. Robots are gradually becoming a popular teaching tool, as robots provide a tangible way for young children to access to technology, and controlling a robot through programming offers them opportunities to engage in developing computational thinking. This study explores whether the introduction of flowcharts into the robotics programming courses can help children convert natural language into a programming language more easily, and then to better cultivate their computational thinking skills. An experimental study was adopted with a sample of children ages six to seven (N = 16) participated, and a one-meter-tall humanoid robot was used as the teaching tool. Results show that children can master basic programming concepts through robotic courses. Children's computational thinking has been significantly improved. Besides, results suggest that flowcharts do have an impact on young children’s computational thinking skills development, but it only has a significant effect on the "sequencing" and "correspondence" skills. Overall, the study demonstrates that the humanoid robot and flowcharts have qualities that foster young children to learn programming and develop computational thinking skills.

Keywords: robotics, computational thinking, programming, young children, flow chart

Procedia PDF Downloads 147

Authors: M. C. Paliwal, A. K. Jain, S. K. Katiyar

Abstract:

Classification of satellite imagery is very important for the assessment of its accuracy. In order to determine the accuracy of the classified image, usually the assumed-true data are derived from ground truth data using Global Positioning System. The data collected from satellite imagery and ground truth data is then compared to find out the accuracy of data and error matrices are prepared. Overall and individual accuracies are calculated using different methods. The study illustrates advanced classification and accuracy assessment of land use/land cover mapping using satellite imagery. IRS-1C-LISS IV data were used for classification of satellite imagery. The satellite image was classified using the software in fourteen classes namely water bodies, agricultural fields, forest land, urban settlement, barren land and unclassified area etc. Classification of satellite imagery and calculation of accuracy was done by using ERDAS-Imagine software to find out the best method. This study is based on the data collected for Bhopal city boundaries of Madhya Pradesh State of India.

Keywords: resolution, accuracy assessment, land use mapping, satellite imagery, ground truth data, error matrices

Procedia PDF Downloads 508

Authors: Joseph George, Anne Kotteswara Roa

Abstract:

Skin disease is one of the most common and popular kinds of health issues faced by people nowadays. Skin cancer (SC) is one among them, and its detection relies on the skin biopsy outputs and the expertise of the doctors, but it consumes more time and some inaccurate results. At the early stage, skin cancer detection is a challenging task, and it easily spreads to the whole body and leads to an increase in the mortality rate. Skin cancer is curable when it is detected at an early stage. In order to classify correct and accurate skin cancer, the critical task is skin cancer identification and classification, and it is more based on the cancer disease features such as shape, size, color, symmetry and etc. More similar characteristics are present in many skin diseases; hence it makes it a challenging issue to select important features from a skin cancer dataset images. Hence, the skin cancer diagnostic accuracy is improved by requiring an automated skin cancer detection and classification framework; thereby, the human expert’s scarcity is handled. Recently, the deep learning techniques like Convolutional neural network (CNN), Deep belief neural network (DBN), Artificial neural network (ANN), Recurrent neural network (RNN), and Long and short term memory (LSTM) have been widely used for the identification and classification of skin cancers. This survey reviews different DL techniques for skin cancer identification and classification. The performance metrics such as precision, recall, accuracy, sensitivity, specificity, and F-measures are used to evaluate the effectiveness of SC identification using DL techniques. By using these DL techniques, the classification accuracy increases along with the mitigation of computational complexities and time consumption.

Keywords: skin cancer, deep learning, performance measures, accuracy, datasets

Procedia PDF Downloads 129

Authors: Akeel A. Shah, Tong Zhang

Abstract:

Computational approaches to learning the properties of materials are commonplace, motivated by the need to screen or design materials for a given application, e.g., semiconductors and energy storage. Experimental approaches can be both time consuming and costly. Unfortunately, computational approaches such as ab-initio electronic structure calculations and classical or ab-initio molecular dynamics are themselves can be too slow for the rapid evaluation of materials, often involving thousands to hundreds of thousands of candidates. Machine learning assisted approaches have been developed to overcome the time limitations of purely physics-based approaches. These approaches, on the other hand, require large volumes of data for training (hundreds of thousands on many standard data sets such as QM7b). This means that they are limited by how quickly such a large data set of physics-based simulations can be established. At high fidelity, such as configuration interaction, composite methods such as G4, and coupled cluster theory, gathering such a large data set can become infeasible, which can compromise the accuracy of the predictions - many applications require high accuracy, for example band structures and energy levels in semiconductor materials and the energetics of charge transfer in energy storage materials. In order to circumvent this problem, multi-fidelity approaches can be adopted, for example the Δ-ML method, which learns a high-fidelity output from a low-fidelity result such as Hartree-Fock or density functional theory (DFT). The general strategy is to learn a map between the low and high fidelity outputs, so that the high-fidelity output is obtained a simple sum of the physics-based low-fidelity and correction, Although this requires a low-fidelity calculation, it typically requires far fewer high-fidelity results to learn the correction map, and furthermore, the low-fidelity result, such as Hartree-Fock or semi-empirical ZINDO, is typically quick to obtain, For high-fidelity outputs the result can be an order of magnitude or more in speed up. In this work, a new multi-fidelity approach is developed, based on a graph convolutional network (GCN) combined with semi-supervised learning. The GCN allows for the material or molecule to be represented as a graph, which is known to improve accuracy, for example SchNet and MEGNET. The graph incorporates information regarding the numbers of, types and properties of atoms; the types of bonds; and bond angles. They key to the accuracy in multi-fidelity methods, however, is the incorporation of low-fidelity output to learn the high-fidelity equivalent, in this case by learning their difference. Semi-supervised learning is employed to allow for different numbers of low and high-fidelity training points, by using an additional GCN-based low-fidelity map to predict high fidelity outputs. It is shown on 4 different data sets that a significant (at least one order of magnitude) increase in accuracy is obtained, using one to two orders of magnitude fewer low and high fidelity training points. One of the data sets is developed in this work, pertaining to 1000 simulations of quinone molecules (up to 24 atoms) at 5 different levels of fidelity, furnishing the energy, dipole moment and HOMO/LUMO.

Keywords: .materials screening, computational materials, machine learning, multi-fidelity, graph convolutional network, semi-supervised learning

Procedia PDF Downloads 41

Authors: Erick C. Valverde, Wansu Lim

Abstract:

Facial emotion recognition (FER) system has been recently developed for more advanced computer vision applications. The ability to identify human emotions would enable smart healthcare facility to diagnose mental health illnesses (e.g., depression and stress) as well as better human social interactions with smart technologies. The FER system involves two steps: 1) face detection task and 2) facial emotion recognition task. It classifies the human expression in various categories such as angry, disgust, fear, happy, sad, surprise, and neutral. This system requires intensive research to address issues with human diversity, various unique human expressions, and variety of human facial features due to age differences. These issues generally affect the ability of the FER system to detect human emotions with high accuracy. Early stage of FER systems used simple supervised classification task algorithms like K-nearest neighbors (KNN) and artificial neural networks (ANN). These conventional FER systems have issues with low accuracy due to its inefficiency to extract significant features of several human emotions. To increase the accuracy of FER systems, deep learning (DL)-based methods, like convolutional neural networks (CNN), are proposed. These methods can find more complex features in the human face by means of the deeper connections within its architectures. However, the inference speed and computational costs of a DL-based FER system is often disregarded in exchange for higher accuracy results. To cope with this drawback, an optimized DL-based FER system is proposed in this study.An extreme version of Inception V3, known as Xception model, is leveraged by applying different network optimization methods. Specifically, network pruning and quantization are used to enable lower computational costs and reduce memory usage, respectively. To support low resource requirements, a 68-landmark face detector from Dlib is used in the early step of the FER system.Furthermore, a DL compiler is utilized to incorporate advanced optimization techniques to the Xception model to improve the inference speed of the FER system. In comparison to VGG-Net and ResNet50, the proposed optimized DL-based FER system experimentally demonstrates the objectives of the network optimization methods used. As a result, the proposed approach can be used to create an efficient and real-time FER system.

Keywords: deep learning, face detection, facial emotion recognition, network optimization methods

Procedia PDF Downloads 118

Authors: Wu Di, Yeo Khoon Seng, Lim Tee Tai

Abstract:

The typical insects employ a flapping-wing mode of flight. The numerical simulations on free flight of a model fruit fly (Re=143) including hovering and are presented in this paper. Unsteady aerodynamics around a flapping insect is studied by solving the three-dimensional Newtonian dynamics of the flyer coupled with Navier-Stokes equations. A hybrid-grid scheme (Generalized Finite Difference Method) that combines great geometry flexibility and accuracy of moving boundary definition is employed for obtaining flow dynamics. The results show good points of agreement and consistency with the outcomes and analyses of other researchers, which validate the computational model and demonstrate the feasibility of this computational approach on analyzing fluid phenomena in insect flight. The present modeling approach also offers a promising route of investigation that could complement as well as overcome some of the limitations of physical experiments in the study of free flight aerodynamics of insects. The results are potentially useful for the design of biomimetic flapping-wing flyers.

Keywords: free hovering flight, flapping wings, fruit fly, insect aerodynamics, leading edge vortex (LEV), computational fluid dynamics (CFD), Navier-Stokes equations (N-S), fluid structure interaction (FSI), generalized finite-difference method (GFD)

Procedia PDF Downloads 410

Authors: Andrey V. Timofeev

Abstract:

This paper introduces an original method for guaranteed estimation of the accuracy of an ensemble of Lipschitz classifiers. The solution was obtained as a finite closed set of alternative hypotheses, which contains an object of classification with a probability of not less than the specified value. Thus, the classification is represented by a set of hypothetical classes. In this case, the smaller the cardinality of the discrete set of hypothetical classes is, the higher is the classification accuracy. Experiments have shown that if the cardinality of the classifiers ensemble is increased then the cardinality of this set of hypothetical classes is reduced. The problem of the guaranteed estimation of the accuracy of an ensemble of Lipschitz classifiers is relevant in the multichannel classification of target events in C-OTDR monitoring systems. Results of suggested approach practical usage to accuracy control in C-OTDR monitoring systems are present.

Keywords: Lipschitz classifiers, confidence set, C-OTDR monitoring, classifiers accuracy, classifiers ensemble

Procedia PDF Downloads 492

Authors: Musaab Riyadh, Norwati Mustapha, Dina Riyadh

Abstract:

Trajectory similarity can be defined as the cost of transforming one trajectory into another based on certain similarity method. It is the core of numerous mining tasks such as clustering, classification, and indexing. Various approaches have been suggested to measure similarity based on the geometric and dynamic properties of trajectory, the overlapping between trajectory segments, and the confined area between entire trajectories. In this article, an evaluation of these approaches has been done based on computational cost, usage memory, accuracy, and the amount of data which is needed in advance to determine its suitability to stream mining applications. The evaluation results show that the stream mining applications support similarity methods which have low computational cost and memory, single scan on data, and free of mathematical complexity due to the high-speed generation of data.

Keywords: global distance measure, local distance measure, semantic trajectory, spatial dimension, stream data mining

Procedia PDF Downloads 396

Authors: Ella Sèdé Maforikan

Abstract:

Accurate land cover mapping is essential for effective environmental monitoring and natural resources management. This study focuses on assessing the classification performance of two satellite datasets and evaluating the impact of different input feature combinations on classification accuracy in the Beterou catchment, situated in the northern part of Benin. Landsat-8 and Sentinel-2 images from June 1, 2020, to March 31, 2021, were utilized. Employing the Random Forest (RF) algorithm on Google Earth Engine (GEE), a supervised classification categorized the land into five classes: forest, savannas, cropland, settlement, and water bodies. GEE was chosen due to its high-performance computing capabilities, mitigating computational burdens associated with traditional land cover classification methods. By eliminating the need for individual satellite image downloads and providing access to an extensive archive of remote sensing data, GEE facilitated efficient model training on remote sensing data. The study achieved commendable overall accuracy (OA), ranging from 84% to 85%, even without incorporating spectral indices and terrain metrics into the model. Notably, the inclusion of additional input sources, specifically terrain features like slope and elevation, enhanced classification accuracy. The highest accuracy was achieved with Sentinel-2 (OA = 91%, Kappa = 0.88), slightly surpassing Landsat-8 (OA = 90%, Kappa = 0.87). This underscores the significance of combining diverse input sources for optimal accuracy in land cover mapping. The methodology presented herein not only enables the creation of precise, expeditious land cover maps but also demonstrates the prowess of cloud computing through GEE for large-scale land cover mapping with remarkable accuracy. The study emphasizes the synergy of different input sources to achieve superior accuracy. As a future recommendation, the application of Light Detection and Ranging (LiDAR) technology is proposed to enhance vegetation type differentiation in the Beterou catchment. Additionally, a cross-comparison between Sentinel-2 and Landsat-8 for assessing long-term land cover changes is suggested.

Keywords: land cover mapping, Google Earth Engine, random forest, Beterou catchment

Procedia PDF Downloads 63

Authors: Rafid A. Al-Khannak, Fawzi M. Al-Naima

Abstract:

Power load is one of the most important controlling keys which decide power demands and illustrate power usage to shape power market. Hence, power load forecasting is the parameter which facilitates understanding and analyzing all these aspects. In this paper, power load forecasting is solved under MATLAB environment by constructing a neural network for the power load to find an accurate simulated solution with the minimum error. A developed algorithm to achieve load forecasting application with faster technique is the aim for this paper. The algorithm is used to enable MATLAB power application to be implemented by multi machines in the Grid computing system, and to accomplish it within much less time, cost and with high accuracy and quality. Grid Computing, the modern computational distributing technology, has been used to enhance the performance of power applications by utilizing idle and desired Grid contributor(s) by sharing computational power resources.

Keywords: DeskGrid, Grid Server, idle contributor(s), grid computing, load forecasting

Procedia PDF Downloads 475

Authors: Hongbo Zhang, Xinlu Tang, Jiangwei Li, Chi Yan

Abstract:

Home zone parking pilot (HPP) is a fast-growing segment in low-speed autonomous driving applications. It requires the car automatically cruise around a parking lot and park itself in a range of up to 100 meters inside a recurrent home/office parking lot, which requires precise parking lot mapping and localization solution. Although Lidar is ideal for SLAM, the car OEMs favor a low-cost fish-eye camera based visual SLAM approach. Recent approaches have employed segmentation models to extract semantic features and improve mapping accuracy, but these AI models are memory unfriendly and computationally expensive, making deploying on embedded ADAS systems difficult. To address this issue, we proposed a new method that utilizes object detection models to extract robust and accurate parking lot features. The proposed method could reduce computational costs while maintaining high accuracy. Once combined with vehicles’ wheel-pulse information, the system could construct maps and locate the vehicle in real-time. This article will discuss in detail (1) the fish-eye based Around View Monitoring (AVM) with transparent chassis images as the inputs, (2) an Object Detection (OD) based feature point extraction algorithm to generate point cloud, (3) a low computational parking lot mapping algorithm and (4) the real-time localization algorithm. At last, we will demonstrate the experiment results with an embedded ADAS system installed on a real car in the underground parking lot.

Keywords: ADAS, home zone parking pilot, object detection, visual SLAM

Procedia PDF Downloads 67

Authors: Malo Pocheau-Lesteven, Olivier Le Maître

Abstract:

Engineering and optimisation of complex systems is often achieved through multi-disciplinary analysis of the system, where each subsystem is modeled and interacts with other subsystems to model the complete system. The coherence of the output of the different sub-systems is achieved through the use of compatibility constraints, which enforce the coupling between the different subsystems. Due to the complexity of some sub-systems and the computational cost of evaluating their respective models, it is often necessary to build surrogate models of these subsystems to allow repeated evaluation these subsystems at a relatively low computational cost. In this paper, gaussian processes are used, as their probabilistic nature is leveraged to evaluate the likelihood of satisfying the compatibility constraints. This paper presents infilling strategies to build accurate surrogate models of the subsystems in areas where they are likely to meet the compatibility constraint. It is shown that these infilling strategies can reduce the computational cost of building surrogate models for a given level of accuracy. An application of these methods to velocity prediction programs used in offshore racing naval architecture further demonstrates these method's applicability in a real engineering context. Also, some examples of the application of uncertainty quantification to field of naval architecture are presented.

Keywords: infilling strategy, gaussian process, multi disciplinary analysis, velocity prediction program

Procedia PDF Downloads 157

Authors: Cristian P. Topa, Esteban A. Valencia, Victor H. Hidalgo, Marco A. Martinez

Abstract:

Wind tunnel experiments for aerodynamic profiles display numerous advantages, such as: clean steady laminar flow, controlled environmental conditions, streamlines visualization, and real data acquisition. However, the experiment instrumentation usually is expensive, and hence, each test implies a incremented in design cost. The aim of this work is to select and implement a low-cost static pressure data acquisition system for a NACA 2412 airfoil in an open cycle wind tunnel. This work compares wind tunnel experiment with Computational Fluid Dynamics (CFD) simulation and parametric analysis. The experiment was evaluated at Reynolds of 1.65 e⁵, with increasing angles from -5° to 15°. The comparison between the approaches show good enough accuracy, between the experiment and CFD, additional parametric analysis results differ widely from the other methods, which complies with the lack of accuracy of the lateral approach due its simplicity.

Keywords: wind tunnel, low cost instrumentation, experimental testing, CFD simulation

Procedia PDF Downloads 180

Authors: N. Dahbi, M. Daoudi

Abstract:

This paper deals with the use of computational method based on Monte Carlo simulation in order to investigate the transport phenomena of the electron in HgCdTe narrow band gap semiconductor. Via this method we can evaluate the time dependence of the transport parameters: velocity, energy and mobility of electrons through matter (HgCdTe).

Keywords: Monte Carlo, transport parameters, HgCdTe, computational mechanics

Procedia PDF Downloads 475

Authors: Shuo Liu, Dan Liu

Abstract:

In the field of named entity recognition, most of the research has been conducted around simple entities. However, for nested named entities, which still contain entities within entities, it has been difficult to identify them accurately due to their boundary ambiguity. In this paper, a hierarchical recognition model is constructed based on the grammatical structure and semantic features of Chinese text for boundary calculation based on lexical features. The analysis is carried out at different levels in terms of granularity, semantics, and lexicality, respectively, avoiding repetitive work to reduce computational effort and using the semantic features of words to calculate the boundaries of entities to improve the accuracy of the recognition work. The results of the experiments carried out on web-based microblogging data show that the model achieves an accuracy of 86.33% and an F1 value of 89.27% in recognizing nested named entities, making up for the shortcomings of some previous recognition models and improving the efficiency of recognition of nested named entities.

Keywords: coarse-grained, nested named entity, Chinese natural language processing, word embedding, T-SNE dimensionality reduction algorithm

Procedia PDF Downloads 128