Search results for: accuracy

Authors: Bin Li, Mei Liu

Abstract:

Arrival time information is a crucial component of advanced public transport system (APTS). The advertisement of arrival time at stops can help reduce the waiting time and anxiety of passengers, and improve the quality of service. In this research, an experiment was conducted to compare the performance on prediction accuracy and precision between the link-based and the path-based historical travel time based model with the automatic vehicle location (AVL) data collected from an actual bus route. The research results show that the path-based model is superior to the link-based model, and achieves the best improvement on peak hours.

Keywords: bus transit, arrival time prediction, link-based, path-based

Procedia PDF Downloads 359

Authors: Ibrahim Mahamid, A. Al-Ghonamy, M. Aichouni

Abstract:

This study is conducted to identify causes of cost deviations in public construction projects in Saudi Arabia from contractors’ perspective. 41 factors that might affect cost estimating accuracy were identified through literature review and discussion with some construction experts. The factors were tabulated in a questionnaire form and a field survey included 51 contractors from the Northern Province of Saudi Arabia was performed. The results show that the top five important causes are: wrong estimation method, long period between design and time of implementation, cost of labor, cost of machinary and absence of construction-cost data.

Keywords: cost deviation, public construction, cost estimating, Saudi Arabia, contractors

Procedia PDF Downloads 475

Authors: Md. Khayrul Bashar, Md. Saiful Islam, Kimiko Yamashita, Yano Midori

Abstract:

Despite having some progress in human authentication, conventional biometrics (e.g., facial features, fingerprints, retinal scans, gait, voice patterns) are not robust against falsification because they are neither confidential nor secret to an individual. As a non-invasive tool, electrocardiogram (ECG) has recently shown a great potential in human recognition due to its unique rhythms characterizing the variability of human heart structures (chest geometry, sizes, and positions). Moreover, ECG has a real-time vitality characteristic that signifies the live signs, which ensure legitimate individual to be identified. However, the detection accuracy of the current ECG-based methods is not sufficient due to a high variability of the individual’s heartbeats at a different instance of time. These variations may occur due to muscle flexure, the change of mental or emotional states, and the change of sensor positions or long-term baseline shift during the recording of ECG signal. In this study, a new method is proposed for human identification, which is based on the extraction of the local roughness of ECG heartbeat signals. First ECG signal is preprocessed using a second order band-pass Butterworth filter having cut-off frequencies of 0.00025 and 0.04. A number of local binary patterns are then extracted by applying a moving neighborhood window along the ECG signal. At each instant of the ECG signal, the pattern is formed by comparing the ECG intensities at neighboring time points with the central intensity in the moving window. Then, binary weights are multiplied with the pattern to come up with the local roughness description of the signal. Finally, histograms are constructed that describe the heartbeat signals of individual subjects in the database. One advantage of the proposed feature is that it does not depend on the accuracy of detecting QRS complex, unlike the conventional methods. Supervised recognition methods are then designed using minimum distance to mean and Bayesian classifiers to identify authentic human subjects. An experiment with sixty (60) ECG signals from sixty adult subjects from National Metrology Institute of Germany (NMIG) - PTB database, showed that the proposed new method is promising compared to a conventional interval and amplitude feature-based method.

Keywords: human identification, ECG biometrics, local roughness patterns, supervised classification

Procedia PDF Downloads 404

Authors: Johnson Oladele Fatokun, I. P. Akpan

Abstract:

In this paper, the one-dimensional time dependent Schrödinger equation is discretized by the method of lines using a second order finite difference approximation to replace the second order spatial derivative. The evolving system of stiff ordinary differential equation (ODE) in time is solved numerically by an L-stable trapezoidal-like integrator. Results show accuracy of relative maximum error of order 10-4 in the interval of consideration. The performance of the method as compared to an existing scheme is considered favorable.

Keywords: Schrodinger’s equation, partial differential equations, method of lines (MOL), stiff ODE, trapezoidal-like integrator

Procedia PDF Downloads 418

Authors: Ahmad Alwosheel, Ahmed Alqaraawi

Abstract:

This work proposes an unsupervised text-independent probabilistic approach to recognize Quran reciter voice. It is an accurate approach that works on real time applications. This approach does not require a prior information about reciter models. It has two phases, where in the training phase the reciters' acoustical features are modeled using Gaussian Mixture Models, while in the testing phase, unlabeled reciter's acoustical features are examined among GMM models. Using this approach, a high accuracy results are achieved with efficient computation time process.

Keywords: Quran, speaker recognition, reciter recognition, Gaussian Mixture Model

Procedia PDF Downloads 380

Authors: Morteza Esfandiari, Shabnam Jabari, Heather MacGrath, David Coleman

Abstract:

Flood is a severe issue in different places in the world as well as the city of Fredericton, New Brunswick, Canada. The downtown area of Fredericton is close to the Saint John River, which is susceptible to flood around May every year. Recently, the frequency of flooding seems to be increased, especially after the fact that the downtown area and surrounding urban/agricultural lands got flooded in two consecutive years in 2018 and 2019. In order to have an explicit vision of flood span and damage to affected areas, it is necessary to use either flood inundation modelling or satellite data. Due to contingent availability and weather dependency of optical satellites, and limited existing data for the high cost of hydrodynamic models, it is not always feasible to rely on these sources of data to generate quality flood maps after or during the catastrophe. Height Above the Nearest Drainage (HAND), a state-of-the-art topo-hydrological index, normalizes the height of a basin based on the relative elevation along with the stream network and specifies the gravitational or the relative drainage potential of an area. HAND is a relative height difference between the stream network and each cell on a Digital Terrain Model (DTM). The stream layer is provided through a multi-step, time-consuming process which does not always result in an optimal representation of the river centerline depending on the topographic complexity of that region. HAND is used in numerous case studies with quite acceptable and sometimes unexpected results because of natural and human-made features on the surface of the earth. Some of these features might cause a disturbance in the generated model, and consequently, the model might not be able to predict the flow simulation accurately. We propose to include a previously existing stream layer generated by the province of New Brunswick and benefit from culvert maps to improve the water flow simulation and accordingly the accuracy of HAND model. By considering these parameters in our processing, we were able to increase the accuracy of the model from nearly 74% to almost 92%. The improved model can be used for generating highly accurate flood maps, which is necessary for future urban planning and flood damage estimation without any need for satellite imagery or hydrodynamic computations.

Keywords: HAND, DTM, rapid floodplain, simplified conceptual models

Procedia PDF Downloads 151

Authors: Sevil Igit, Merve Meric, Sarp Erturk

Abstract:

In this paper, it is proposed to improve Daisy descriptor based face recognition using a novel One-Bit Transform (1BT) based pre-registration approach. The 1BT based pre-registration procedure is fast and has low computational complexity. It is shown that the face recognition accuracy is improved with the proposed approach. The proposed approach can facilitate highly accurate face recognition using DAISY descriptor with simple matching and thereby facilitate a low-complexity approach.

Keywords: face recognition, Daisy descriptor, One-Bit Transform, image registration

Procedia PDF Downloads 367

Authors: Calvin Ralph, Edward Archer, Alistair McIlhagger

Abstract:

3D woven textile composites continue to emerge as an advanced material for structural applications and composite manufacture due to their bespoke nature, through thickness reinforcement and near net shape capabilities. When 3D woven preforms are produced, they are in their optimal physical state. As 3D weaving is a dry preforming technology it relies on compression of the preform to achieve the desired composite thickness, fibre volume fraction (Vf) and consolidation. This compression of the preform during manufacture results in changes to its thickness and architecture which can often lead to under-performance or changes of the 3D woven composite. Unlike traditional 2D fabrics, the bespoke nature and variability of 3D woven architectures makes it difficult to know exactly how each 3D preform will behave during processing. Therefore, the focus of this study is to investigate the effect of compression on differing 3D woven architectures in terms of structure, crimp or fibre waviness and thickness as well as analysing the accuracy of available software to predict how 3D woven preforms behave under compression. To achieve this, 3D preforms are modelled and compression simulated in Wisetex with varying architectures of binder style, pick density, thickness and tow size. These architectures have then been woven with samples dry compression tested to determine the compressibility of the preforms under various pressures. Additional preform samples were manufactured using Resin Transfer Moulding (RTM) with varying compressive force. Composite samples were cross sectioned, polished and analysed using microscopy to investigate changes in architecture and crimp. Data from dry fabric compression and composite samples were then compared alongside the Wisetex models to determine accuracy of the prediction and identify architecture parameters that can affect the preform compressibility and stability. Results indicate that binder style/pick density, tow size and thickness have a significant effect on compressibility of 3D woven preforms with lower pick density allowing for greater compression and distortion of the architecture. It was further highlighted that binder style combined with pressure had a significant effect on changes to preform architecture where orthogonal binders experienced highest level of deformation, but highest overall stability, with compression while layer to layer indicated a reduction in fibre crimp of the binder. In general, simulations showed a relative comparison to experimental results; however, deviation is evident due to assumptions present within the modelled results.

Keywords: 3D woven composites, compression, preforms, textile composites

Procedia PDF Downloads 135

Authors: Sameh Abaza, Mohamed Ibrahim, Tarek Mahmoud

Abstract:

Due to the development in the digital image processing, its wide use in many applications such as medical, security, and others, the need for more accurate techniques that are reliable, fast and robust is vehemently demanded. In the field of security, in particular, speed is of the essence. In this paper, a pattern recognition technique that is based on the use of Hidden Markov Model (HMM), K-means and the Sobel operator method is developed. The proposed technique is proved to be fast with respect to some other techniques that are investigated for comparison. Moreover, it shows its capability of recognizing the normal face (center part) as well as face boundary.

Keywords: HMM, K-Means, Sobel, accuracy, face recognition

Procedia PDF Downloads 331

Authors: Ava Dodd, Jonathan Adams

Abstract:

The purpose of this study is to explore the characteristics of developing a machine learning application using synthetic data. The study is structured to develop the application for the purpose of deploying the computer vision model. The findings discuss the realities of attempting to develop a computer vision model for practical purpose, and detail the processes, tools, and techniques that were used to meet accuracy requirements. The research reveals that synthetic data represents another variable that can be adjusted to improve the performance of a computer vision model. Further, a suite of tools and tuning recommendations are provided.

Keywords: computer vision, machine learning, synthetic data, YOLOv4

Procedia PDF Downloads 225

Authors: Misuk Kim, Seokho Kang, Jehyuk Lee, Hyunchang Cho, Sungzoon Cho

Abstract:

In semiconductor manufacturing, virtual metrology (VM) refers to methods to predict properties of a wafer based on machine parameters and sensor data of the production equipment, without performing the (costly) physical measurement of the wafer properties (Wikipedia). Additional benefits include avoidance of human bias and identification of important factors affecting the quality of the process which allow improving the process quality in the future. It is however rare to find VM applied to other areas of manufacturing. In this work, we propose to use VM to copper clad laminate (CCL) manufacturing. CCL is a core element of a printed circuit board (PCB) which is used in smartphones, tablets, digital cameras, and laptop computers. The manufacturing of CCL consists of three processes: Treating, lay-up, and pressing. Treating, the most important process among the three, puts resin on glass cloth, heat up in a drying oven, then produces prepreg for lay-up process. In this process, three important quality factors are inspected: Treated weight (T/W), Minimum Viscosity (M/V), and Gel Time (G/T). They are manually inspected, incurring heavy cost in terms of time and money, which makes it a good candidate for VM application. We developed prediction models of the three quality factors T/W, M/V, and G/T, respectively, with process variables, raw material, and environment variables. The actual process data was obtained from a CCL manufacturer. A variety of variable selection methods and learning algorithms were employed to find the best prediction model. We obtained prediction models of M/V and G/T with a high enough accuracy. They also provided us with information on “important” predictor variables, some of which the process engineers had been already aware and the rest of which they had not. They were quite excited to find new insights that the model revealed and set out to do further analysis on them to gain process control implications. T/W did not turn out to be possible to predict with a reasonable accuracy with given factors. The very fact indicates that the factors currently monitored may not affect T/W, thus an effort has to be made to find other factors which are not currently monitored in order to understand the process better and improve the quality of it. In conclusion, VM application to CCL’s treating process was quite successful. The newly built quality prediction model allowed one to reduce the cost associated with actual metrology as well as reveal some insights on the factors affecting the important quality factors and on the level of our less than perfect understanding of the treating process.

Keywords: copper clad laminate, predictive modeling, quality control, virtual metrology

Procedia PDF Downloads 350

Authors: Duckyong Kim, Jong Kang Park, Jong Tae Kim

Abstract:

DOA (Direction of Arrival) estimation is an important method in array signal processing and has a wide range of applications such as direction finding, beam forming, and so on. In this paper, we briefly introduce the MUSIC (Multiple Signal Classification) Algorithm, one of DOA estimation methods for analyzing several targets. Then we apply the MUSIC algorithm to the two-dimensional antenna array to analyze DOA estimation in 3D space through MATLAB simulation. We also analyze the design factors that can affect the accuracy of DOA estimation through simulation, and proceed with further consideration on how to apply the system.

Keywords: DOA estimation, MUSIC algorithm, spatial spectrum, array signal processing

Procedia PDF Downloads 379

Authors: Shaan Khosla, Jon Krohn

Abstract:

In a recruiting context, machine learning models are valuable for recommendations: to predict the best candidates for a vacancy, to match the best vacancies for a candidate, and compile a set of similar candidates for any given candidate. While useful to create these models, validating their accuracy in a recommendation context is difficult due to a sparsity of data. In this report, we use network graph data to generate useful representations for candidates and vacancies. We use candidates and vacancies as network nodes and designate a bi-directional link between them based on the candidate interviewing for the vacancy. After using node2vec, the embeddings are used to construct a validation dataset with a ranked order, which will help validate new recommender systems.

Keywords: AI, machine learning, NLP, recruiting

Procedia PDF Downloads 84

Authors: Wenya Shu, Ilinca Stanciulescu

Abstract:

Carbon nanotubes (CNTs) possess attractive properties, such as high stiffness and strength, and high thermal and electrical conductivities, making them promising filler in multifunctional nanocomposites. Although CNTs can be efficient reinforcements, the expected level of mechanical performance of CNT-polymers is not often reached in practice due to the poor mechanical behavior of the CNT-polymer interfaces. It is believed that the interactions of CNT and polymer mainly result from the Van der Waals force. The interface debonding is a fracture and delamination phenomenon. Thus, the cohesive zone modeling (CZM) is deemed to give good capture of the interface behavior. The detailed, cohesive zone modeling provides an option to consider the CNT-matrix interactions, but brings difficulties in mesh generation and also leads to high computational costs. Homogenized models that smear the fibers in the ground matrix and treat the material as homogeneous are studied in many researches to simplify simulations. But based on the perfect interface assumption, the traditional homogenized model obtained by mixing rules severely overestimates the stiffness of the composite, even comparing with the result of the CZM with artificially very strong interface. A mechanical model that can take into account the interface debonding and achieve comparable accuracy to the CZM is thus essential. The present study first investigates the CNT-matrix interactions by employing cohesive zone modeling. Three different coupled CZM laws, i.e., bilinear, exponential and polynomial, are considered. These studies indicate that the shapes of the CZM constitutive laws chosen do not influence significantly the simulations of interface debonding. Assuming a bilinear traction-separation relationship, the debonding process of single CNT in the matrix is divided into three phases and described by differential equations. The analytical solutions corresponding to these phases are derived. A homogenized model is then developed by introducing a parameter characterizing interface sliding into the mixing theory. The proposed mechanical model is implemented in FEAP8.5 as a user material. The accuracy and limitations of the model are discussed through several numerical examples. The CZM simulations in this study reveal important factors in the modeling of CNT-matrix interactions. The analytical solutions and proposed homogenized model provide alternative methods to efficiently investigate the mechanical behaviors of CNT/polymer composites.

Keywords: carbon nanotube, cohesive zone modeling, homogenized model, interface debonding

Procedia PDF Downloads 129

Authors: Ardalan Sabamehr, Ashutosh Bagchi

Abstract:

Recently, Structural Health Monitoring (SHM) based on the vibration of structures has attracted the attention of researchers in different fields such as: civil, aeronautical and mechanical engineering. Operational Modal Analysis (OMA) have been developed to identify modal properties of infrastructure such as bridge, building and so on. Frequency Domain Decomposition (FDD), Stochastic Subspace Identification (SSI) and Continuous Wavelet Transform (CWT) are the three most common methods in output only modal identification. FDD, SSI, and CWT operate based on the frequency domain, time domain, and time-frequency plane respectively. So, FDD and SSI are not able to display time and frequency at the same time. By the way, FDD and SSI have some difficulties in a noisy environment and finding the closed modes. CWT technique which is currently developed works on time-frequency plane and a reasonable performance in such condition. The other advantage of wavelet transform rather than other current techniques is that it can be applied for the non-stationary signal as well. The aim of this paper is to compare three most common modal identification techniques to find modal properties (such as natural frequency, mode shape, and damping ratio) of three story steel frame which was built in Concordia University Lab by use of ambient vibration. The frame has made of Galvanized steel with 60 cm length, 27 cm width and 133 cm height with no brace along the long span and short space. Three uniaxial wired accelerations (MicroStarin with 100mv/g accuracy) have been attached to the middle of each floor and gateway receives the data and send to the PC by use of Node Commander Software. The real-time monitoring has been performed for 20 seconds with 512 Hz sampling rate. The test is repeated for 5 times in each direction by hand shaking and impact hammer. CWT is able to detect instantaneous frequency by used of ridge detection method. In this paper, partial derivative ridge detection technique has been applied to the local maxima of time-frequency plane to detect the instantaneous frequency. The extracted result from all three methods have been compared, and it demonstrated that CWT has the better performance in term of its accuracy in noisy environment. The modal parameters such as natural frequency, damping ratio and mode shapes are identified from all three methods.

Keywords: ambient vibration, frequency domain decomposition, stochastic subspace identification, continuous wavelet transform

Procedia PDF Downloads 296

Authors: Luisa Mesquita Oliveira Ribeiro, Alexei Manso Correa Machado

Abstract:

Handwritten signature is a unique form for recognizing an individual, used to discern documents, carry out investigations in the criminal, legal, banking areas and other applications. Signature verification is based on large amounts of biometric data, as they are simple and easy to acquire, among other characteristics. Given this scenario, signature forgery is a worldwide recurring problem and fast and precise techniques are needed to prevent crimes of this nature from occurring. This article carried out a study on the efficiency of the Capsule Network in analyzing and recognizing signatures. The chosen architecture achieved an accuracy of 98.11% and 80.15% for the CEDAR and GPDS databases, respectively.

Keywords: biometrics, deep learning, handwriting, signature forgery

Procedia PDF Downloads 83

Authors: Hany Elsaid Fawaz Abdallah

Abstract:

This study presents a new artificial neural network(ANN) model to predict the Nusselt Number and pressure drop for the turbulent flow in a triangular corrugated plate heat exchanger for forced air and turbulent water flow. An experimental investigation was performed to create a new dataset for the Nusselt Number and pressure drop values in the following range of dimensionless parameters: The plate corrugation angles (from 0° to 60°), the Reynolds number (from 10000 to 40000), pitch to height ratio (from 1 to 4), and Prandtl number (from 0.7 to 200). Based on the ANN performance graph, the three-layer structure with {12-8-6} hidden neurons has been chosen. The training procedure includes back-propagation with the biases and weight adjustment, the evaluation of the loss function for the training and validation dataset and feed-forward propagation of the input parameters. The linear function was used at the output layer as the activation function, while for the hidden layers, the rectified linear unit activation function was utilized. In order to accelerate the ANN training, the loss function minimization may be achieved by the adaptive moment estimation algorithm (ADAM). The ‘‘MinMax’’ normalization approach was utilized to avoid the increase in the training time due to drastic differences in the loss function gradients with respect to the values of weights. Since the test dataset is not being used for the ANN training, a cross-validation technique is applied to the ANN network using the new data. Such procedure was repeated until loss function convergence was achieved or for 4000 epochs with a batch size of 200 points. The program code was written in Python 3.0 using open-source ANN libraries such as Scikit learn, TensorFlow and Keras libraries. The mean average percent error values of 9.4% for the Nusselt number and 8.2% for pressure drop for the ANN model have been achieved. Therefore, higher accuracy compared to the generalized correlations was achieved. The performance validation of the obtained model was based on a comparison of predicted data with the experimental results yielding excellent accuracy.

Keywords: artificial neural networks, corrugated channel, heat transfer enhancement, Nusselt number, pressure drop, generalized correlations

Procedia PDF Downloads 87

Authors: Byung hyun Bae

Abstract:

In power transformer, leakage magnetic flux generate temperature rise of inner steel plate. Sometimes, this temperature rise can be serious problem. If temperature of steel plate is over critical point, harmful gas will be generated in the tank. And this gas can be a reason of fire, explosion and life decrease. So, temperature rise forecasting of steel plate is very important at the design stage of power transformer. To improve accuracy of forecasting of temperature rise, comparison between simulation and experiment achieved in this paper.

Keywords: power transformer, steel plate, temperature rise, experiment, simulation

Procedia PDF Downloads 495

Authors: Nateepat Srivarom, Weng Jingnong, Serm Chinnarat

Abstract:

Precise Point Positioning (PPP) technique is use to improve accuracy by using precise satellite orbit and clock correction data, but this technique is complicated methods and high costs. Currently, there are several online processing service providers which offer simplified calculation. In the first part of this research, we compare the efficiency and precision of four software. There are three popular online processing service providers: Australian Online GPS Processing Service (AUSPOS), CSRS-Precise Point Positioning and CenterPoint RTX post processing by Trimble and 1 offline software, RTKLIB, which collected data from 10 the International GNSS Service (IGS) stations for 10 days. The results indicated that AUSPOS has the least distance root mean square (DRMS) value of 0.0029 which is good enough to be calculated for monitoring the movement of tectonic plates. The second, we use AUSPOS to process the data of geodetic network of Thailand. In December 26, 2004, the earthquake occurred a 9.3 MW at the north of Sumatra that highly affected all nearby countries, including Thailand. Earthquake effects have led to errors of the coordinate system of Thailand. The Royal Thai Survey Department (RTSD) is primarily responsible for monitoring of the crustal movement of the country. The difference of the geodetic network movement is not the same network and relatively large. This result is needed for survey to continue to improve GPS coordinates system in every year. Therefore, in this research we chose the AUSPOS to calculate the magnitude and direction of movement, to improve coordinates adjustment of the geodetic network consisting of 19 pins in Thailand during October 2013 to November 2017. Finally, results are displayed on the simulation map by using the ArcMap program with the Inverse Distance Weighting (IDW) method. The pin with the maximum movement is pin no. 3239 (Tak) in the northern part of Thailand. This pin moved in the south-western direction to 11.04 cm. Meanwhile, the directional movement of the other pins in the south gradually changed from south-west to south-east, i.e., in the direction noticed before the earthquake. The magnitude of the movement is in the range of 4 - 7 cm, implying small impact of the earthquake. However, the GPS network should be continuously surveyed in order to secure accuracy of the geodetic network of Thailand.

Keywords: precise point positioning, online processing service, geodetic network, inverse distance weighting

Procedia PDF Downloads 189

Authors: Ahmed R. Abdelaziz, Zaki M. I. Osha

Abstract:

Solar radiation data are of great significance for solar energy system design. This study aims at developing and calibrating new empirical models for estimating monthly mean daily global solar radiation on a horizontal surface in Alexandria, Egypt. Day length hours, sun height, day number, and declination angle calculated data are used for this purpose. A comparison between measured and calculated values of solar radiation is carried out. It is shown that all the proposed correlations are able to predict the global solar radiation with excellent accuracy in Alexandria.

Keywords: solar energy, global solar radiation, model, regression coefficient

Procedia PDF Downloads 405

Authors: Archit Yajnik

Abstract:

Part of Speech Tagging has always been a challenging task in the era of Natural Language Processing. This article presents POS tagging for Nepali text using Hidden Markov Model and Viterbi algorithm. From the Nepali text, annotated corpus training and testing data set are randomly separated. Both methods are employed on the data sets. Viterbi algorithm is found to be computationally faster and accurate as compared to HMM. The accuracy of 95.43% is achieved using Viterbi algorithm. Error analysis where the mismatches took place is elaborately discussed.

Keywords: hidden markov model, natural language processing, POS tagging, viterbi algorithm

Procedia PDF Downloads 327

Authors: Tolulope Aremu

Abstract:

The key process steps to produce liquid detergent products will introduce potential defects, such as formulation, mixing, filling, and packaging, which might compromise product quality, consumer safety, and operational efficiency. Real-time identification and characterization of such defects are of prime importance for maintaining high standards and reducing waste and costs. Usually, defect detection is performed by human inspection or rule-based systems, which is very time-consuming, inconsistent, and error-prone. The present study overcomes these limitations in dealing with optimization in defect characterization within the process for making liquid detergents using Machine Learning algorithms. Performance testing of various machine learning models was carried out: Support Vector Machine, Decision Trees, Random Forest, and Convolutional Neural Network on defect detection and classification of those defects like wrong viscosity, color deviations, improper filling of a bottle, packaging anomalies. These algorithms have significantly benefited from a variety of optimization techniques, including hyperparameter tuning and ensemble learning, in order to greatly improve detection accuracy while minimizing false positives. Equipped with a rich dataset of defect types and production parameters consisting of more than 100,000 samples, our study further includes information from real-time sensor data, imaging technologies, and historic production records. The results are that optimized machine learning models significantly improve defect detection compared to traditional methods. Take, for instance, the CNNs, which run at 98% and 96% accuracy in detecting packaging anomaly detection and bottle filling inconsistency, respectively, by fine-tuning the model with real-time imaging data, through which there was a reduction in false positives of about 30%. The optimized SVM model on detecting formulation defects gave 94% in viscosity variation detection and color variation. These values of performance metrics correspond to a giant leap in defect detection accuracy compared to the usual 80% level achieved up to now by rule-based systems. Moreover, this optimization with models can hasten defect characterization, allowing for detection time to be below 15 seconds from an average of 3 minutes using manual inspections with real-time processing of data. With this, the reduction in time will be combined with a 25% reduction in production downtime because of proactive defect identification, which can save millions annually in recall and rework costs. Integrating real-time machine learning-driven monitoring drives predictive maintenance and corrective measures for a 20% improvement in overall production efficiency. Therefore, the optimization of machine learning algorithms in defect characterization optimum scalability and efficiency for liquid detergent companies gives improved operational performance to higher levels of product quality. In general, this method could be conducted in several industries within the Fast moving consumer Goods industry, which would lead to an improved quality control process.

Keywords: liquid detergent manufacturing, defect detection, machine learning, support vector machines, convolutional neural networks, defect characterization, predictive maintenance, quality control, fast-moving consumer goods

Procedia PDF Downloads 18

Authors: Chung-Ming Lo, Chung-Chien Lee

Abstract:

In many countries, the lifetime prevalence of shoulder pain is up to 70%. In America, the health care system spends 7 billion per year about the healthy issues of shoulder pain. With respect to the origin, up to 70% of shoulder pain is attributed to rotator cuff lesions This study proposed a computer-aided diagnosis (CAD) system to assist radiologists classifying rotator cuff lesions with less operator dependence. Quantitative features were extracted from the shoulder ultrasound images acquired using an ALOKA alpha-6 US scanner (Hitachi-Aloka Medical, Tokyo, Japan) with linear array probe (scan width: 36mm) ranging from 5 to 13 MHz. During examination, the postures of the examined patients are standard sitting position and are followed by the regular routine. After acquisition, the shoulder US images were drawn out from the scanner and stored as 8-bit images with pixel value ranging from 0 to 255. Upon the sonographic appearance, the boundary of each lesion was delineated by a physician to indicate the specific pattern for analysis. The three lesion categories for classification were composed of 20 cases of tendon inflammation, 18 cases of calcific tendonitis, and 18 cases of supraspinatus tear. For each lesion, second-order statistics were quantified in the feature extraction. The second-order statistics were the texture features describing the correlations between adjacent pixels in a lesion. Because echogenicity patterns were expressed via grey-scale. The grey-scale co-occurrence matrixes with four angles of adjacent pixels were used. The texture metrics included the mean and standard deviation of energy, entropy, correlation, inverse different moment, inertia, cluster shade, cluster prominence, and Haralick correlation. Then, the quantitative features were combined in a multinomial logistic regression classifier to generate a prediction model of rotator cuff lesions. Multinomial logistic regression classifier is widely used in the classification of more than two categories such as the three lesion types used in this study. In the classifier, backward elimination was used to select a feature subset which is the most relevant. They were selected from the trained classifier with the lowest error rate. Leave-one-out cross-validation was used to evaluate the performance of the classifier. Each case was left out of the total cases and used to test the trained result by the remaining cases. According to the physician’s assessment, the performance of the proposed CAD system was shown by the accuracy. As a result, the proposed system achieved an accuracy of 86%. A CAD system based on the statistical texture features to interpret echogenicity values in shoulder musculoskeletal ultrasound was established to generate a prediction model for rotator cuff lesions. Clinically, it is difficult to distinguish some kinds of rotator cuff lesions, especially partial-thickness tear of rotator cuff. The shoulder orthopaedic surgeon and musculoskeletal radiologist reported greater diagnostic test accuracy than general radiologist or ultrasonographers based on the available literature. Consequently, the proposed CAD system which was developed according to the experiment of the shoulder orthopaedic surgeon can provide reliable suggestions to general radiologists or ultrasonographers. More quantitative features related to the specific patterns of different lesion types would be investigated in the further study to improve the prediction.

Keywords: shoulder ultrasound, rotator cuff lesions, texture, computer-aided diagnosis

Procedia PDF Downloads 284

Authors: J. K. Alhassan, C. S. Actsu

Abstract:

This study is on the prediction of feature performance of undergraduate students with Artificial Neural Networks (ANN). With the growing decline in the quality academic performance of undergraduate students, it has become essential to predict the students’ feature academic performance early in their courses of first and second years and to take the necessary precautions using such prediction-based information. The feed forward multilayer neural network model was used to train and develop a network and the test carried out with some of the input variables. A result of 80% accuracy was obtained from the test which was carried out, with an average error of 0.009781.

Keywords: academic performance, artificial neural network, prediction, students

Procedia PDF Downloads 467

Authors: A. Shaygani, R. Saifi, M. S. Saidi, M. Sani

Abstract:

Particle size distribution, the most important characteristics of aerosols, is obtained through electrical characterization techniques. The dynamics of charged nano-particles under the influence of electric field in electrical mobility spectrometer (EMS) reveals the size distribution of these particles. The accuracy of this measurement is influenced by flow conditions, geometry, electric field and particle charging process, therefore by the transfer function (transfer matrix) of the instrument. In this work, a wire-cylinder corona charger was designed and the combined field-diffusion charging process of injected poly-disperse aerosol particles was numerically simulated as a prerequisite for the study of a multi-channel EMS. The result, a cloud of particles with non-uniform charge distribution, was introduced to the EMS. The flow pattern and electric field in the EMS were simulated using computational fluid dynamics (CFD) to obtain particle trajectories in the device and therefore to calculate the reported signal by each electrometer. According to the output signals (resulted from bombardment of particles and transferring their charges as currents), we proposed a modification to the size of detecting rings (which are connected to electrometers) in order to evaluate particle size distributions more accurately. Based on the capability of the system to transfer information contents about size distribution of the injected particles, we proposed a benchmark for the assessment of optimality of the design. This method applies the concept of Von Neumann entropy and borrows the definition of entropy from information theory (Shannon entropy) to measure optimality. Entropy, according to the Shannon entropy, is the ''average amount of information contained in an event, sample or character extracted from a data stream''. Evaluating the responses (signals) which were obtained via various configurations of detecting rings, the best configuration which gave the best predictions about the size distributions of injected particles, was the modified configuration. It was also the one that had the maximum amount of entropy. A reasonable consistency was also observed between the accuracy of the predictions and the entropy content of each configuration. In this method, entropy is extracted from the transfer matrix of the instrument for each configuration. Ultimately, various clouds of particles were introduced to the simulations and predicted size distributions were compared to the exact size distributions.

Keywords: aerosol nano-particle, CFD, electrical mobility spectrometer, von neumann entropy

Procedia PDF Downloads 343

Authors: Shao-Tzu Huang, Chen-Chien Hsu, Wei-Yen Wang

Abstract:

Matching high dimensional features between images is computationally expensive for exhaustive search approaches in computer vision. Although the dimension of the feature can be degraded by simplifying the prior knowledge of homography, matching accuracy may degrade as a tradeoff. In this paper, we present a feature matching method based on k-means algorithm that reduces the matching cost and matches the features between images instead of using a simplified geometric assumption. Experimental results show that the proposed method outperforms the previous linear exhaustive search approaches in terms of the inlier ratio of matched pairs.

Keywords: feature matching, k-means clustering, SIFT, RANSAC

Procedia PDF Downloads 357

Authors: Chin-Yu Lin, Kun-Chi Wang, Shih-Hsu Wang, Wei-Chih Wang

Abstract:

A construction superintendent needs to know not only the amount of quantities of cost items or materials completed to develop a daily report or calculate the daily progress (earned value) in each day, but also the amount of quantities of materials (e.g., reinforced steel and concrete) to be ordered (or moved into the jobsite) for performing the in-progress or ready-to-start construction activities (e.g., erection of reinforced steel and concrete pouring). These daily construction management tasks require great effort in extracting accurate quantities in a short time (usually must be completed right before getting off work every day). As a result, most superintendents can only provide these quantity data based on either what they see on the site (high inaccuracy) or the extraction of quantities from two-dimension (2D) construction drawings (high time consumption). Hence, the current practice of providing the amount of quantity data completed in each day needs improvement in terms of more accuracy and efficiency. Recently, a three-dimension (3D)-based building information model (BIM) technique has been widely applied to support construction quantity takeoffs (QTO) process. The capability of virtual reality (VR) allows to view a building from the first person's viewpoint. Thus, this study proposes an innovative system by integrating VR (using 'Unity') and BIM (using 'Revit') to extract quantities to support the above daily construction management tasks. The use of VR allows a system user to be present in a virtual building to more objectively assess the construction progress in the office. This VR- and BIM-based system is also facilitated by an integrated database (consisting of the information and data associated with the BIM model, QTO, and costs). In each day, a superintendent can work through a BIM-based virtual building to quickly identify (via a developed VR shooting function) the building components (or objects) that are in-progress or finished in the jobsite. And he then specifies a percentage (e.g., 20%, 50% or 100%) of completion of each identified building object based on his observation on the jobsite. Next, the system will generate the completed quantities that day by multiplying the specified percentage by the full quantities of the cost items (or materials) associated with the identified object. A building construction project located in northern Taiwan is used as a case study to test the benefits (i.e., accuracy and efficiency) of the proposed system in quantity extraction for supporting the development of daily reports and the orders of construction materials.

Keywords: building information model, construction management, quantity takeoffs, virtual reality

Procedia PDF Downloads 132

Authors: Xuezhang Hou

Abstract:

In the present paper, a hybrid hyperbolic dynamic system formulated by partial differential equations with initial and boundary conditions is considered. First, the system is transformed to an abstract evolution system in an appropriate Hilbert space, and spectral analysis and semigroup generation of the system operator is discussed. Subsequently, a sliding model control problem is proposed and investigated, and an equivalent control method is introduced and applied to the system. Finally, a significant result that the state of the system can be approximated by an ideal sliding mode under control in any accuracy is derived and examined.

Keywords: hyperbolic dynamic system, sliding model control, semigroup of linear operators, partial differential equations

Procedia PDF Downloads 136

Authors: Elyes Zarrouk, Yassine Benayed, Faiez Gargouri

Abstract:

This paper presents a new hybrid model for isolated Arabic words recognition. To do this, we apply Support Vectors Machine (SVM) as an estimator of posterior probabilities within the Dynamic Bayesian networks (DBN). This paper deals a comparative study between DBN and SVM/DBN systems for multi-dialect isolated Arabic words. Performance using SVM/DBN is found to exceed that of DBNs trained on an identical task, giving higher recognition accuracy for four different Arabic dialects. In fact, the average of recognition rates for the four dialects with SVM/DBN was 87.67% while 83.01% with DBN.

Keywords: dynamic Bayesian networks, hybrid models, supports vectors machine, Arabic isolated words

Procedia PDF Downloads 560

Authors: Narumol Chintaganun

Abstract:

In this paper we present the second-order central differencing scheme with the staggered version (STG) for solving the advection equation and Burger's equation. This scheme based on staggered evolution of the re-constructed cell averages. This scheme results in the second-order central differencing scheme, an extension along the lines of the first-order central scheme of Lax-Friedrichs (LxF) scheme. All numerical simulations presented in this paper are obtained by finite difference method (FDM) and STG. Numerical results are shown that the STG gives very good results and higher accuracy.

Keywords: central differencing scheme, STG, advection equation, burgers equation

Procedia PDF Downloads 557