Search results for: dimensional accuracy (DA)

Authors: Degale Desta, Cheng Jian

Abstract:

The most important factor in identifying a person is their face. Even identical twins have their own distinct faces. As a result, identification and face recognition are needed to tell one person from another. A face recognition system is a verification tool used to establish a person's identity using biometrics. Nowadays, face recognition is a common technique used in a variety of applications, including home security systems, criminal identification, and phone unlock systems. This system is more secure because it only requires a facial image instead of other dependencies like a key or card. Face detection and face identification are the two phases that typically make up a human recognition system.The idea behind designing and creating a face recognition system using deep learning with Azure ML Python's OpenCV is explained in this paper. Face recognition is a task that can be accomplished using deep learning, and given the accuracy of this method, it appears to be a suitable approach. To show how accurate the suggested face recognition system is, experimental results are given in 98.46% accuracy using Fast-RCNN Performance of algorithms under different training conditions.

Keywords: deep learning, face recognition, identification, fast-RCNN

Procedia PDF Downloads 140

Authors: Zohreh Derikvand, Hadis Cheraghi, Azadeh Azadbakht, Vaclav Eigner, Michal Dusek

Abstract:

Two new one and two dimensional metal organic coordination polymers of Cu(II), [Cu(3-nph)2(H2O)2pz]n (1) and Ag(I), {[Ag(3-nph)pz].H2O}n (2) with pyrazine (pz) and 3- nitrophthalic acid (3-nph) have been synthesized and characterized by elemental analysis, spectral (IR, UV-Vis), thermal (TG/DTG) analysis and single crystal X-ray diffraction. We used these compounds to preparation modified electrode with Au/Pt nanosparticles in order to investigation electrochemistry and electrocatalysis activities. The surface structure and composition of the sensor were characterized by scanning electron microscopy (SEM). The Ag(I) coordination polymer shows a 2D layer structure constructed from dinuclear silver (I) building blocks in which two crystallographically Ag+ ions are connected to each other by a covalent bond. The pyrazine ligands adopt μ2 bridging modes, linking the metal centers into a one and two -dimensional coordination framework in 1 and 2. The two AgI cations are surrounded by pyrazine and 3-nitrophthalate mono anions and indicate distorted tetrahedral geometry. In the crystal structures of Ag(I) complex there are non-classical hydrogen bonding arrangements, C–O•••π and π–π stacking interactions. In Cu(II) coordination polymer, the coordination geometry around Cu(II) atom is a distorted octahedron. Interestingly, the structural analysis illustrates that the strong and weak hydrogen bond accompanied with C–H•••π and C–O•••π stacking interactions assemble the crystal structure of 1 and 2 into fascinating 3D supramolecular architecture.

Keywords: 3-nithrophethalic acid, crystal structure, coordination polymer, electrocatalysis

Procedia PDF Downloads 319

Authors: Muhammet Baldan, Emel Timuçin

Abstract:

Molecular solubility prediction plays a crucial role in various fields, such as drug discovery, environmental science, and material science. In this study, we compare the performance of five machine learning algorithms—linear regression, support vector machines (SVM), random forests, gradient boosting machines (GBM), and neural networks—for predicting molecular solubility using the AqSolDB dataset. The dataset consists of 9981 data points with their corresponding solubility values. MACCS keys (166 bits), RDKit properties (20 properties), and structural properties(3) features are extracted for every smile representation in the dataset. A total of 189 features were used for training and testing for every molecule. Each algorithm is trained on a subset of the dataset and evaluated using metrics accuracy scores. Additionally, computational time for training and testing is recorded to assess the efficiency of each algorithm. Our results demonstrate that random forest model outperformed other algorithms in terms of predictive accuracy, achieving an 0.93 accuracy score. Gradient boosting machines and neural networks also exhibit strong performance, closely followed by support vector machines. Linear regression, while simpler in nature, demonstrates competitive performance but with slightly higher errors compared to ensemble methods. Overall, this study provides valuable insights into the performance of machine learning algorithms for molecular solubility prediction, highlighting the importance of algorithm selection in achieving accurate and efficient predictions in practical applications.

Keywords: random forest, machine learning, comparison, feature extraction

Procedia PDF Downloads 41

Authors: Hassan Athari, Abdurrahim Bolukbasi, Dogan Ciloglu

Abstract:

This study represented the respiratory alveoli system, and determined the trajectory of inhaled particles more accurately using the modified three-dimensional model with deformable walls of alveoli. The study also considered the tissue tension in the model to demonstrate the effect of lung. Tissue tensions are transferred by the lung parenchyma and produce the pressure gradient. This load expands the alveoli and establishes a sub-ambient (vacuum) pressure within the lungs. Thus, at the alveolar level, the flow field and movement of alveoli wall lead to an integrated effect. In this research, we assume that the three-dimensional alveolus has a visco-elastic tissue (walls). For accurate investigation of pulmonary tissue mechanical properties on particle transport and alveolar flow field, the actual relevance between tissue movement and airflow is solved by two-way FSI (Fluid Structure Interaction) simulation technique in the alveolus. Therefore, the essence of real simulation of pulmonary breathing mechanics can be achieved by developing a coupled FSI computational model. We, therefore conduct a series of FSI simulations over a range of tissue models and breathing rates. As a result, the fluid flows and streamlines have changed during present flexible model against the rigid models and also the two-way coupling particle trajectories have changed against the one-way particle coupling.

Keywords: FSI, two-way particle coupling, alveoli, CDF

Procedia PDF Downloads 258

Authors: Guorong Sui, Xingwei Jia, Fei Tong, Xiumin Gao

Abstract:

Camera calibration is a fundamental issue of high precision noncontact measurement. And it is necessary to analyze and study the reliability and application range of its linear model which is often used in the camera calibration. According to the imaging features of monocular cameras, a camera model which is based on the image pixel coordinates and three dimensional space coordinates is built. Using our own customized template, the image pixel coordinate is obtained by the subpixel corner detection method. Without considering the aberration of the optical system, the feature extraction and linearity analysis of the line segment in the template are performed. Moreover, the experiment is repeated 11 times by constantly varying the measuring distance. At last, the linearity of the camera is achieved by fitting 11 groups of data. The camera model measurement results show that the relative error does not exceed 1%, and the repeated measurement error is not more than 0.1 mm magnitude. Meanwhile, it is found that the model has some measurement differences in the different region and object distance. The experiment results show this linear model is simple and practical, and have good linearity within a certain object distance. These experiment results provide a powerful basis for establishment of the linear model of camera. These works will have potential value to the actual engineering measurement.

Keywords: camera linear model, geometric imaging relationship, image pixel coordinates, three dimensional space coordinates, sub-pixel corner detection

Procedia PDF Downloads 277

Authors: Nikolai A. Khlebnikov, Vladimir N. Krasilnikov, Evgenii V. Polyakov, Anastasia A. Maltceva

Abstract:

Carbon materials with advanced surfaces are widely used both in modern industry and in environmental protection. The physical-chemical nature of these materials is determined by the morphology of primary atomic and molecular carbon structures, which are the basis for synthesizing the following materials: zero-dimensional (fullerenes), one-dimensional (fiber, tubes), two-dimensional (graphene) carbon nanostructures, three-dimensional (multi-layer graphene, graphite, foams) with unique physical-chemical and functional properties. Experience shows that the microscopic morphological level is the basis for the creation of the next mesoscopic morphological level. The dependence of the morphology on the chemical way and process prehistory (crystallization, colloids formation, liquid crystal state and other) is the peculiarity of the last called level. These factors determine the consumer properties of carbon materials, such as specific surface area, porosity, chemical resistance in corrosive environments, catalytic and adsorption activities. Based on the developed ideology of thin precursor synthesis, the authors discuss one of the approaches of the porosity control of carbon-containing materials with a given aggregates morphology. The low-temperature thermolysis of precursors in a gas environment of a given composition is the basis of the above-mentioned idea. The processes of carbothermic precursor synthesis of two different compounds: tungsten carbide WC:nC and zinc oxide ZnO:nC containing an impurity phase in the form of free carbon were selected as subjects of the research. In the first case, the transition metal (tungsten) forming carbides was the object of the synthesis. In the second case, there was selected zinc that does not form carbides. The synthesis of both kinds of transition metals compounds was conducted by the method of precursor carbothermic synthesis from the organic solution. ZnO:nC composites were obtained by thermolysis of succinate Zn(OO(CH2)2OO), formate glycolate Zn(HCOO)(OCH2CH2O)1/2, glycerolate Zn(OCH2CHOCH2OH), and tartrate Zn(OOCCH(OH)CH(OH)COO). WC:nC composite was synthesized from ammonium paratungstate and glycerol. In all cases, carbon structures that are specific for diamond- like carbon forms appeared on the surface of WC and ZnO particles after the heat treatment. Tungsten carbide and zinc oxide were removed from the composites by selective chemical dissolution preserving the amorphous carbon phase. This work presents the results of investigating WC:nC and ZnO:nC composites and carbon nanopowders with tubular, tape, plate and onion morphologies of aggregates that are separated by chemical dissolution of WC and ZnO from the composites by the following methods: SEM, TEM, XPA, Raman spectroscopy, and BET. The connection between the carbon morphology under the conditions of synthesis and chemical nature of the precursor and the possibility of regulation of the morphology with the specific surface area up to 1700-2000 m2/g of carbon-structured materials are discussed.

Keywords: carbon morphology, composite materials, precursor synthesis, tungsten carbide, zinc oxide

Procedia PDF Downloads 335

Authors: Mohammed Bnebaghdad, Feriel Betouche, Malika Semmani

Abstract:

Image processing become more sophisticated over time due to technological advances, especially artificial intelligence (AI) technology. Currently, AI image processing is used in many areas, including surveillance, industry, science, and medicine. AI in medical image processing can help doctors diagnose diseases faster, with minimal mistakes, and with less effort. Among these diseases is malaria, which remains a major public health challenge in many parts of the world. It affects millions of people every year, particularly in tropical and subtropical regions. Early detection of malaria is essential to prevent serious complications and reduce the burden of the disease. In this paper, we propose and implement a scheme based on AI image processing to enhance malaria disease diagnosis through automated analysis of blood smear images. The scheme is based on the convolutional neural network (CNN) method. So, we have developed a model that classifies infected and uninfected single red cells using images available on Kaggle, as well as real blood smear images obtained from the Central Laboratory of Medical Biology EHS Laadi Flici (formerly El Kettar) in Algeria. The real images were segmented into individual cells using the watershed algorithm in order to match the images from the Kaagle dataset. The model was trained and tested, achieving an accuracy of 99% and 97% accuracy for new real images. This validates that the model performs well with new real images, although with slightly lower accuracy. Additionally, the model has been embedded in a Raspberry Pi4, and a graphical user interface (GUI) was developed to visualize the malaria diagnostic results and facilitate user interaction.

Keywords: medical image processing, malaria parasite, classification, CNN, artificial intelligence

Procedia PDF Downloads 20

Authors: Yuqing Chen, Ying Xu, Renfa Li

Abstract:

The development of matrix completion theory provides new approaches for data gathering in Wireless Sensor Networks (WSN). The existing matrix completion algorithms for WSN mainly consider how to reduce the sampling number without considering the real-time performance when recovering the data matrix. In order to guarantee the recovery accuracy and reduce the recovery time consumed simultaneously, we propose a new ALM algorithm to recover the weather monitoring data. A lot of experiments have been carried out to investigate the performance of the proposed ALM algorithm by using different parameter settings, different sampling rates and sampling models. In addition, we compare the proposed ALM algorithm with some existing algorithms in the literature. Experimental results show that the ALM algorithm can obtain better overall recovery accuracy with less computing time, which demonstrate that the ALM algorithm is an effective and efficient approach for recovering the real world weather monitoring data in WSN.

Keywords: wireless sensor network, matrix completion, singular value thresholding, augmented Lagrange multiplier

Procedia PDF Downloads 384

Authors: Dua Hişam, Serhat İkizoğlu

Abstract:

Identifying the problem behind balance disorder is one of the most interesting topics in the medical literature. This study has considerably enhanced the development of artificial intelligence (AI) algorithms applying multiple machine learning (ML) models to sensory data on gait collected from humans to classify between normal people and those suffering from Vestibular System (VS) problems. Although AI is widely utilized as a diagnostic tool in medicine, AI models have not been used to perform feature extraction and identify VS disorders through training on raw data. In this study, three machine learning (ML) models, the Random Forest Classifier (RF), Extreme Gradient Boosting (XGB), and K-Nearest Neighbor (KNN), have been trained to detect VS disorder, and the performance comparison of the algorithms has been made using accuracy, recall, precision, and f1-score. With an accuracy of 95.28 %, Random Forest Classifier (RF) was the most accurate model.

Keywords: vestibular disorder, machine learning, random forest classifier, k-nearest neighbor, extreme gradient boosting

Procedia PDF Downloads 69

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: Shoukat Ali, Amjad Hussain, Latif-ur-Rehman, Sehrish Inam

Abstract:

Radiotherapy plays an important role in the management of cancer patients. Approximately 50% of the cancer patients receive radiotherapy at one point or another during the course of treatment. The entire radiotherapy treatment of curative intent is divided into different phases, depending on the histology of the tumor. The established protocols are useful in deciding the total dose, fraction size, and numbers of phases. The objective of this study was to evaluate the dosimetric differences between the conventional treatment protocols and the three-dimensional conformal simultaneously integrated boost (SIB) plans for three different tumors sites (i.e. bladder, breast, and brain). A total of 30 patients with brain, breast and bladder cancers were selected in this retrospective study. All the patients were CT simulated initially. The primary physician contoured PTV1 and PTV2 in the axial slices. The conventional doses prescribed for brain and breast is 60Gy/30 fractions, and 64.8Gy/36 fractions for bladder treatment. For the SIB plans biological effective doses (BED) were calculated for 25 fractions. The two conventional (Phase I and Phase II) and a single SIB plan for each patient were generated on Eclipse™ treatment planning system. Treatment plans were compared and analyzed for coverage index, conformity index, homogeneity index, dose gradient and organs at risk doses.In both plans 95% of PTV volume received a minimum of 95% of the prescribe dose. Dose deviation in the optic chiasm was found to be less than 0.5%. There is no significant difference in lung V20 and heart V30 in the breast plans. In the rectum plans V75%, V50% and V25% were found to be less than 1.2% different. Deviation in the tumor coverage, conformity and homogeneity indices were found to be less than 1%. SIB plans with three dimensional conformal radiotherapy technique reduce the overall treatment time without compromising the target coverage and without increasing dose to the organs at risk. The higher dose per fraction may increase the late effects to some extent. Further studies are required to evaluate the late effects with the intention of standardizing the SIB technique for practical implementation.

Keywords: coverage index, conformity index, dose gradient, homogeneity index, simultaneously integrated boost

Procedia PDF Downloads 476

Authors: Sina Fatolahzadeh, Sergio A. Sepúlveda

Abstract:

This paper presents advanced developments in close-range, terrestrial remote sensing techniques to enhance the characterization of rock masses. The study integrates two state-of-the-art laser-scanning technologies, the HandySCAN and GeoSLAM laser scanners, to extract high-resolution geospatial data for rock mass analysis. These instruments offer high accuracy, precision, low acquisition time, and high efficiency in capturing intricate geological features in small to medium size outcrops and slope cuts. Using the HandySCAN and GeoSLAM laser scanners facilitates real-time, three-dimensional mapping of rock surfaces, enabling comprehensive assessments of rock mass characteristics. The collected data provide valuable insights into structural complexities, surface roughness, and discontinuity patterns, which are essential for geological and geotechnical analyses. The synergy of these advanced remote sensing technologies contributes to a more precise and straightforward understanding of rock mass behavior. In this case, the main parameters of RQD, joint spacing, persistence, aperture, roughness, infill, weathering, water condition, and joint orientation in a slope cut along the Sea-to-Sky Highway, BC, were remotely analyzed to calculate and evaluate the Rock Mass Rating (RMR) and Geological Strength Index (GSI) classification systems. Automatic and manual analyses of the acquired data are then compared with field measurements. The results show the usefulness of the proposed remote sensing methods and their appropriate conformity with the actual field data.

Keywords: remote sensing, rock mechanics, rock engineering, slope stability, discontinuity properties

Procedia PDF Downloads 66

Authors: Anchana Julsiri, Seree Chadcham

Abstract:

The present study determined the effects of listening to pleasant Thai classical music on increasing working memory in elderly. Thai classical music without lyrics that made participants feel fun and aroused was used in the experiment for 3.19-5.40 minutes. The accuracy scores of Counting Span Task (CST), upper alpha ERD%, and theta ERS% were used to assess working memory of participants both before and after listening to pleasant Thai classical music. The results showed that the accuracy scores of CST and upper alpha ERD% in the frontal area of participants after listening to Thai classical music were significantly higher than before listening to Thai classical music (p < .05). Theta ERS% in the fronto-parietal network of participants after listening to Thai classical music was significantly lower than before listening to Thai classical music (p < .05).

Keywords: brain wave, elderly, pleasant Thai classical music, working memory

Procedia PDF Downloads 459

Authors: Andrey V. Timofeev, Dmitry V. Egorov

Abstract:

The practical efficient approach is suggested for estimation of the seismoacoustic sources energy in C-OTDR monitoring systems. This approach represents the sequential plan for confidence estimation both the seismoacoustic sources energy, as well the absorption coefficient of the soil. The sequential plan delivers the non-asymptotic guaranteed accuracy of obtained estimates in the form of non-asymptotic confidence regions with prescribed sizes. These confidence regions are valid for a finite sample size when the distributions of the observations are unknown. Thus, suggested estimates are non-asymptotic and nonparametric, and also these estimates guarantee the prescribed estimation accuracy in the form of the prior prescribed size of confidence regions, and prescribed confidence coefficient value.

Keywords: nonparametric estimation, sequential confidence estimation, multichannel monitoring systems, C-OTDR-system, non-lineary regression

Procedia PDF Downloads 357

Authors: Bhim Kumar Dahal

Abstract:

Transportation network development in the developing country is in rapid pace. The majority of the network belongs to railway and expressway which passes through diverse topography, landform and geological conditions despite the avoidance principle during route selection. Construction of such networks demand many low to high embankment which required improvement in the foundation soil. This paper is mainly focused on the various advanced ground improvement techniques used to improve the soft soil, modelling approach and its predictability for embankments construction. The ground improvement techniques can be broadly classified in to three groups i.e. densification group, drainage and consolidation group and reinforcement group which are discussed with some case studies.  Various methods were used in modelling of the embankments from simple 1-dimensional to complex 3-dimensional model using variety of constitutive models. However, the reliability of the predictions is not found systematically improved with the level of sophistication.  And sometimes the predictions are deviated more than 60% to the monitored value besides using same level of erudition. This deviation is found mainly due to the selection of constitutive model, assumptions made during different stages, deviation in the selection of model parameters and simplification during physical modelling of the ground condition. This deviation can be reduced by using optimization process, optimization tools and sensitivity analysis of the model parameters which will guide to select the appropriate model parameters.

Keywords: cement, improvement, physical properties, strength

Procedia PDF Downloads 174

Authors: Bharatendra Rai

Abstract:

Predictive data analysis and modeling involving machine learning techniques become challenging in presence of too many explanatory variables or features. Presence of too many features in machine learning is known to not only cause algorithms to slow down, but they can also lead to decrease in model prediction accuracy. This study involves housing dataset with 79 quantitative and qualitative features that describe various aspects people consider while buying a new house. Boruta algorithm that supports feature selection using a wrapper approach build around random forest is used in this study. This feature selection process leads to 49 confirmed features which are then used for developing predictive random forest models. The study also explores five different data partitioning ratios and their impact on model accuracy are captured using coefficient of determination (r-square) and root mean square error (rsme).

Keywords: housing data, feature selection, random forest, Boruta algorithm, root mean square error

Procedia PDF Downloads 323

Authors: Li Kewen, Su Zhaoxin, Wang Xingmou, Zhu Jian Bing

Abstract:

Conventional reservoir prediction methods ar not sufficient to explore the implicit relation between seismic attributes, and thus data utilization is low. In order to improve the predictive classification accuracy of reservoir lithology, this paper proposes a deep learning lithology prediction method based on ResNet (Residual Neural Network) and SENet (Squeeze-and-Excitation Neural Network). The neural network model is built and trained by using seismic attribute data and lithology data of Shengli oilfield, and the nonlinear mapping relationship between seismic attribute and lithology marker is established. The experimental results show that this method can significantly improve the classification effect of reservoir lithology, and the classification accuracy is close to 70%. This study can effectively predict the lithology of undrilled area and provide support for exploration and development.

Keywords: convolutional neural network, lithology, prediction of reservoir, seismic attributes

Procedia PDF Downloads 177

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

Authors: Jacques Cuenca, Claudio Colangeli, Agnieszka Mroz, Karl Janssens, Gunther Riexinger, Antonio D'Antuono, Giuseppe Pandarese, Milena Martarelli, Gian Marco Revel, Carlos Barcena Martin

Abstract:

This study presents an on-site acoustic inspection methodology for quality and performance evaluation of building components. The work focuses on global and detailed sound source localisation, by successively performing acoustic beamforming and sound intensity measurements. A portable experimental setup is developed, consisting of an omnidirectional broadband acoustic source and a microphone array and sound intensity probe. Three main acoustic indicators are of interest, namely the sound pressure distribution on the surface of components such as walls, windows and junctions, the three-dimensional sound intensity field in the vicinity of junctions, and the sound transmission loss of partitions. The measurement data is post-processed and converted into a three-dimensional numerical model of the acoustic indicators with the help of the simultaneously acquired geolocation information. The three-dimensional acoustic indicators are then integrated into an augmented reality platform superimposing them onto a real-time visualisation of the spatial environment. The methodology thus enables a measurement-supported inspection process of buildings and the correction of errors during construction and refurbishment. Two experimental validation cases are shown. The first consists of a laboratory measurement on a full-scale mockup of a room, featuring a prefabricated panel. The latter is installed with controlled defects such as lack of insulation and joint sealing material. It is demonstrated that the combined acoustic and augmented reality tool is capable of identifying acoustic leakages from the building defects and assist in correcting them. The second validation case is performed on a prefabricated room at a near-completion stage in the factory. With the help of the measurements and visualisation tools, the homogeneity of the partition installation is evaluated and leakages from junctions and doors are identified. Furthermore, the integration of acoustic indicators together with thermal and geometrical indicators via the augmented reality platform is shown.

Keywords: acoustic inspection, prefabricated building components, augmented reality, sound source localization

Procedia PDF Downloads 384

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 290

Authors: Rahul Patel

Abstract:

Ahmedabad, a city located in western India, is experiencing rapid growth due to urbanization and industrialization. It is projected to become a metropolitan city in the near future, resulting in various construction activities. Soil testing is necessary before construction can commence, requiring construction companies and contractors to periodically conduct soil testing. The focus of this study is on the process of creating a spatial database that is digitally formatted and integrated with geotechnical data and a Geographic Information System (GIS). Building a comprehensive geotechnical (Geo)-database involves three steps: collecting borehole data from reputable sources, verifying the accuracy and redundancy of the data, and standardizing and organizing the geotechnical information for integration into the database. Once the database is complete, it is integrated with GIS, allowing users to visualize, analyze, and interpret geotechnical information spatially. Using a Topographic to Raster interpolation process in GIS, estimated values are assigned to all locations based on sampled geotechnical data values. The study area was contoured for SPT N-Values, Soil Classification, Φ-Values, and Bearing Capacity (T/m2). Various interpolation techniques were cross-validated to ensure information accuracy. This GIS map enables the calculation of SPT N-Values, Φ-Values, and bearing capacities for different footing widths and various depths. This study highlights the potential of GIS in providing an efficient solution to complex phenomena that would otherwise be tedious to achieve through other means. Not only does GIS offer greater accuracy, but it also generates valuable information that can be used as input for correlation analysis. Furthermore, this system serves as a decision support tool for geotechnical engineers.

Keywords: ArcGIS, borehole data, geographic information system, geo-database, interpolation, SPT N-value, soil classification, Φ-Value, bearing capacity

Procedia PDF Downloads 74

Authors: Belkacem Laimouche

Abstract:

With the field of artificial intelligence (AI) experiencing exponential growth, fueled by technological advancements that pave the way for increasingly innovative and promising applications, there is an escalating need to develop rigorous methods for assessing their performance in pursuit of transparency and equity. This article proposes a metrology-inspired statistical framework for evaluating bias and explainability in AI systems. Drawing from the principles of metrology, we propose a pioneering approach, using a concrete example, to evaluate the accuracy and precision of AI models, as well as to quantify the sources of measurement uncertainty that can lead to bias in their predictions. Furthermore, we explore a statistical approach for evaluating the explainability of AI systems based on their ability to provide interpretable and transparent explanations of their predictions.

Keywords: artificial intelligence, metrology, measurement uncertainty, prediction error, bias, machine learning algorithms, probabilistic models, interlaboratory comparison, data analysis, data reliability, measurement of bias impact on predictions, improvement of model accuracy and reliability

Procedia PDF Downloads 105

Authors: Milad Vahidi, Mahmod R. Sahebi, Mehrnoosh Omati, Reza Mohammadi

Abstract:

Forest management organizations need information to perform their work effectively. Remote sensing is an effective method to acquire information from the Earth. Two datasets of remote sensing images were used to classify forested regions. Firstly, all of extractable features from hyperspectral and PolSAR images were extracted. The optical features were spectral indexes related to the chemical, water contents, structural indexes, effective bands and absorption features. Also, PolSAR features were the original data, target decomposition components, and SAR discriminators features. Secondly, the particle swarm optimization (PSO) and the genetic algorithms (GA) were applied to select optimization features. Furthermore, the support vector machine (SVM) classifier was used to classify the image. The results showed that the combination of PSO and SVM had higher overall accuracy than the other cases. This combination provided overall accuracy about 90.56%. The effective features were the spectral index, the bands in shortwave infrared (SWIR) and the visible ranges and certain PolSAR features.

Keywords: hyperspectral, PolSAR, feature selection, SVM

Procedia PDF Downloads 416

Authors: Junyu Chen, Peng Xu

Abstract:

In the context of global efforts to enhance building energy efficiency, accurate thermal load forecasting is crucial for both device sizing and predictive control. Variable Refrigerant Flow (VRF) systems are widely used in buildings around the world, yet VRF zonal load prediction has received limited attention. Due to differences between VRF zones in building-level prediction methods, zone-level load forecasting could significantly enhance accuracy. Given that modern VRF systems generate high-quality data, this paper introduces transfer learning to leverage this data and further improve prediction performance. This framework also addresses the challenge of predicting load for building zones with no historical data, offering greater accuracy and usability compared to pure white-box models. The study first establishes an initial variable set of VRF zonal building loads and generates a foundational white-box database using EnergyPlus. Key variables for VRF zonal loads are identified using methods including SRRC, PRCC, and Random Forest. XGBoost and LSTM are employed to generate pre-trained black-box models based on the white-box database. Finally, real-world data is incorporated into the pre-trained model using transfer learning to enhance its performance in operational buildings. In this paper, zone-level load prediction was integrated with transfer learning, and a framework was proposed to improve the accuracy and applicability of VRF zonal load prediction.

Keywords: zonal load prediction, variable refrigerant flow (VRF) system, transfer learning, energyplus

Procedia PDF Downloads 28

Authors: Mahshid Arabi

Abstract:

In recent years, remarkable advancements in the fields of artificial intelligence and machine learning have led to the development of facial recognition technologies. These technologies are now employed in a wide range of applications, including security, surveillance, healthcare, and education. In the field of education, the identification of university entrance exam candidates has been one of the fundamental challenges. Traditional methods such as using ID cards and handwritten signatures are not only inefficient and prone to fraud but also susceptible to errors. In this context, utilizing advanced technologies like facial recognition can be an effective and efficient solution to increase the accuracy and reliability of identity verification in entrance exams. This article examines the use of FaceMatch software for recognizing the faces of university entrance exam candidates in Iran. The main objective of this research is to evaluate the efficiency and accuracy of FaceMatch software in identifying university entrance exam candidates to prevent fraud and ensure the authenticity of individuals' identities. Additionally, this research investigates the advantages and challenges of using this technology in Iran's educational systems. This research was conducted using an experimental method and random sampling. In this study, 1000 university entrance exam candidates in Iran were selected as samples. The facial images of these candidates were processed and analyzed using FaceMatch software. The software's accuracy and efficiency were evaluated using various metrics, including accuracy rate, error rate, and processing time. The research results indicated that FaceMatch software could accurately identify candidates with a precision of 98.5%. The software's error rate was less than 1.5%, demonstrating its high efficiency in facial recognition. Additionally, the average processing time for each candidate's image was less than 2 seconds, indicating the software's high efficiency. Statistical evaluation of the results using precise statistical tests, including analysis of variance (ANOVA) and t-test, showed that the observed differences were significant, and the software's accuracy in identity verification is high. The findings of this research suggest that FaceMatch software can be effectively used as a tool for identifying university entrance exam candidates in Iran. This technology not only enhances security and prevents fraud but also simplifies and streamlines the exam administration process. However, challenges such as preserving candidates' privacy and the costs of implementation must also be considered. The use of facial recognition technology with FaceMatch software in Iran's educational systems can be an effective solution for preventing fraud and ensuring the authenticity of university entrance exam candidates' identities. Given the promising results of this research, it is recommended that this technology be more widely implemented and utilized in the country's educational systems.

Keywords: facial recognition, FaceMatch software, Iran, university entrance exam

Procedia PDF Downloads 48

Authors: Mitsuyoshi Tomiya, Kentaro Kawamura, Shoichi Sakamoto

Abstract:

In chaotic billiard systems, scar-like localization has been found on time-evolving wave packet. We may call it the “dynamical scar” to separate it to the original scar in stationary states. It also comes out along the vicinity of classical unstable periodic orbits, when the wave packets are launched along the orbits, against the hypothesis that the waves become homogenous all around the billiard. Then time-evolving wave packets are investigated numerically in phase space. The Wigner function is adopted to detect the wave packets in phase space. The 2-dimensional Poincaré sections of the 4-dimensional phase space are introduced to clarify the dynamical behavior of the wave packets. The Poincaré sections of the coordinate (x or y) and the momentum (Px or Py) can visualize the dynamical behavior of the wave packets, including the behavior in the momentum degree also. For example, in “dynamical scar” states, a bit larger momentum component comes first, and then the a bit smaller and smaller components follow next. The sections made in the momentum space (Px or Py) elucidates specific trajectories that have larger contribution to the “dynamical scar” states. It is the fixed point observation of the momentum degrees at a specific fixed point(x0, y0) in the phase space. The accumulation are also calculated to search the “dynamical scar” in the Poincare sections. It is found the scars as bright spots in momentum degrees of the phase space.

Keywords: chaotic billiard, Poincaré section, scar, wave packet

Procedia PDF Downloads 452

Authors: Norhidayu Muhamad Zain

Abstract:

Gelatin, a purified protein derived mostly from porcine and bovine sources, is used widely in food manufacturing, pharmaceutical, and cosmetic industries. However, the presence of any porcine-related products are strictly forbidden for Muslim and Jewish consumption. Therefore, analytical methods offering reliable results to differentiate the sources of gelatin are needed. The aim of this study was to differentiate the sources of gelatin (porcine and tilapia fish) using Fourier transform infrared- attenuated total reflection (FTIR-ATR) combined with two dimensional infrared (2DIR) correlation analysis. Porcine gelatin (PG) and tilapia fish gelatin (FG) samples were diluted in distilled water at concentrations ranged from 4-20% (w/v). The samples were then analysed using FTIR-ATR and 2DIR correlation software. The results showed a significant difference in the pattern map of synchronous spectra at the region of 1000 cm⁻¹ to 1100 cm⁻¹ between PG and FG samples. The auto peak at 1080 cm⁻¹ that attributed to C-O functional group was observed at high intensity in PG samples compared to FG samples. Meanwhile, two auto peaks (1080 cm⁻¹ and 1030 cm⁻¹) at lower intensity were identified in FG samples. In addition, using 2D correlation analysis, the original broad water OH bands in 1D IR spectra can be effectively differentiated into six auto peaks located at 3630, 3340, 3230, 3065, 2950 and 2885 cm⁻¹ for PG samples and five auto peaks at 3630, 3330, 3230, 3060 and 2940 cm⁻¹ for FG samples. Based on the rule proposed by Noda, the sequence of the spectral changes in PG samples is as following: NH₃⁺ amino acid > CH₂ and CH₃ aliphatic > OH stretch > carboxylic acid OH stretch > NH in secondary amide > NH in primary amide. In contrast, the sequence was totally in the opposite direction for FG samples and thus both samples provide different 2D correlation spectra ranged from 2800 cm-1 to 3700 cm⁻¹. This method may provide a rapid determination of gelatin source for application in food, pharmaceutical, and cosmetic products.

Keywords: 2 dimensional infrared (2DIR) correlation analysis, Fourier transform infrared- attenuated total reflection (FTIR-ATR), porcine gelatin, tilapia fish gelatin

Procedia PDF Downloads 250

Authors: Fuad Ali Mohammed Al-Yarimi

Abstract:

Many algorithms have been proposed to provide privacy preserving in data mining. These protocols are based on two main approaches named as: the perturbation approach and the Cryptographic approach. The first one is based on perturbation of the valuable information while the second one uses cryptographic techniques. The perturbation approach is much more efficient with reduced accuracy while the cryptographic approach can provide solutions with perfect accuracy. However, the cryptographic approach is a much slower method and requires considerable computation and communication overhead. In this paper, a new scalable protocol is proposed which combines the advantages of the perturbation and distortion along with cryptographic approach to perform privacy preserving in distributed frequent itemset mining on horizontally distributed data. Both the privacy and performance characteristics of the proposed protocol are studied empirically.

Keywords: anonymity data, data mining, distributed frequent itemset mining, gaussian perturbation, perturbation approach, privacy preserving data mining

Procedia PDF Downloads 505

Authors: Shraddha Jain Sharma, Ratnalata Gupta

Abstract:

Predicting stock movements in the financial market is both difficult and rewarding. Analysts and academics are increasingly using advanced approaches such as machine learning techniques to anticipate stock price patterns, thanks to the expanding capacity of computing and the recent advent of graphics processing units and tensor processing units. Stock market prediction is a type of time series prediction that is incredibly difficult to do since stock prices are influenced by a variety of financial, socioeconomic, and political factors. Furthermore, even minor mistakes in stock market price forecasts can result in significant losses for companies that employ the findings of stock market price prediction for financial analysis and investment. Soft computing techniques are increasingly being employed for stock market prediction due to their better accuracy than traditional statistical methodologies. The proposed research looks at the need for soft computing techniques in stock market prediction, the numerous soft computing approaches that are important to the field, past work in the area with their prominent features, and the significant problems or issue domain that the area involves. For constructing a predictive model, the major focus is on neural networks and fuzzy logic. The stock market is extremely unpredictable, and it is unquestionably tough to correctly predict based on certain characteristics. This study provides a complete overview of the numerous strategies investigated for high accuracy prediction, with a focus on the most important characteristics.

Keywords: stock market prediction, artificial intelligence, artificial neural networks, fuzzy logic, accuracy, deep learning, machine learning, stock price, trading volume

Procedia PDF Downloads 90

Authors: Khaled Abou-El-Hossein

Abstract:

The presentation will address the features of two IR convex lenses that have been manufactured using an ultra-high precision machining centre based on single-point diamond turning. The lenses are made from silicon and germanium with a radius of curvature of 500 mm. Because of the brittle nature of silicon and germanium, machining parameters were selected in such a way that ductile regime was achieved. The cutting speed was 800 rpm while the feed rate and depth cut were 20 mm/min and 20 um, respectively. Although both materials comprise a mono-crystalline microstructure and are quite similar in terms of optical properties, machining of silicon was accompanied with more difficulties in terms of form accuracy compared to germanium machining. The P-V error of the silicon profile was 0.222 um while it was only 0.055 um for the germanium lens. This could be attributed to the accelerated wear that takes place on the tool edge when turning mono-crystalline silicon. Currently, we are using other ranges of the machining parameters in order to determine their optimal range that could yield satisfactory performance in terms of form accuracy when fabricating silicon lenses.

Keywords: diamond turning, optical surfaces, precision machining, surface roughness

Procedia PDF Downloads 317