Search results for: fault detection and classification

Authors: Noha Seddik, Sherine Youssef, Mohamed Kholeif

Abstract:

The automated epileptic seizure detection research field has emerged in the recent years; this involves analyzing the Electroencephalogram (EEG) signals instead of the traditional visual inspection performed by expert neurologists. In this study, a Support Vector Machine (SVM) that uses Weighted Permutation Entropy (WPE) as the input feature is proposed for classifying normal and seizure EEG records. WPE is a modified statistical parameter of the permutation entropy (PE) that measures the complexity and irregularity of a time series. It incorporates both the mapped ordinal pattern of the time series and the information contained in the amplitude of its sample points. The proposed system utilizes the fact that entropy based measures for the EEG segments during epileptic seizure are lower than in normal EEG.

Keywords: electroencephalogram (EEG), epileptic seizure detection, weighted permutation entropy (WPE), support vector machine (SVM)

Procedia PDF Downloads 357

Authors: Gulfam Haider, sana danish

Abstract:

Convolutional Neural Networks (CNNs) have emerged as powerful tools for image classification, and the choice of optimizers profoundly affects their performance. The study of optimizers and their adaptations remains a topic of significant importance in machine learning research. While numerous studies have explored and advocated for various optimizers, the efficacy of these optimization techniques is still subject to scrutiny. This work aims to address the challenges surrounding the effectiveness of optimizers by conducting a comprehensive analysis and evaluation. The primary focus of this investigation lies in examining the performance of different optimizers when employed in conjunction with the popular activation function, Rectified Linear Unit (ReLU). By incorporating ReLU, known for its favorable properties in prior research, the aim is to bolster the effectiveness of the optimizers under scrutiny. Specifically, we evaluate the adjustment of these optimizers with both the original Softmax activation function and the modified ReLU activation function, carefully assessing their impact on overall performance. To achieve this, a series of experiments are conducted using a well-established benchmark dataset for image classification tasks, namely the Canadian Institute for Advanced Research dataset (CIFAR-10). The selected optimizers for investigation encompass a range of prominent algorithms, including Adam, Root Mean Squared Propagation (RMSprop), Adaptive Learning Rate Method (Adadelta), Adaptive Gradient Algorithm (Adagrad), and Stochastic Gradient Descent (SGD). The performance analysis encompasses a comprehensive evaluation of the classification accuracy, convergence speed, and robustness of the CNN models trained with each optimizer. Through rigorous experimentation and meticulous assessment, we discern the strengths and weaknesses of the different optimization techniques, providing valuable insights into their suitability for image classification tasks. By conducting this in-depth study, we contribute to the existing body of knowledge surrounding optimizers in CNNs, shedding light on their performance characteristics for image classification. The findings gleaned from this research serve to guide researchers and practitioners in making informed decisions when selecting optimizers and activation functions, thus advancing the state-of-the-art in the field of image classification with convolutional neural networks.

Keywords: deep neural network, optimizers, RMsprop, ReLU, stochastic gradient descent

Procedia PDF Downloads 107

Authors: Newdeskarl Saint Fleur, Nathalie Feuillet, Raphaël Grandin, Éric Jacques, Jennifer Weil-Accardo, Yann Klinger

Abstract:

The prevailing consensus is that the 2010 Mw7.0 Haiti earthquake left the Enriquillo–Plantain Garden strike-slip Fault (EPGF) unruptured but broke unmapped blind north-dipping thrusts. Using high-resolution topography, aerial images, bathymetry and geology we identified previously unrecognized south-dipping NW-SE-striking active thrusts in southern Haiti. One of them, Lamentin thrust (LT), cuts across the crowded city of Carrefour, extends offshore into Port-au-Prince Bay and connects at depth with the EPGF. We propose that both faults broke in 2010. The rupture likely initiated on the thrust and propagated further along the EPGF due to unclamping. This scenario is consistent with geodetic, seismological and field data. The 2010 earthquake increased the stress toward failure on the unruptured segments of the EPGF and on neighboring thrusts, significantly increasing the seismic hazard in the Port-au-Prince urban area. The numerous active thrusts recognized in that area must be considered for future evaluation of the seismic hazard.

Keywords: active faulting, enriquillo-plantain garden fault, Haiti earthquake, seismic hazard

Procedia PDF Downloads 1222

Authors: Ahmed El-Banbi

Abstract:

Several PVT models exist to represent how PVT properties are handled in sub-surface and surface engineering calculations for oil and gas production. The most commonly used models include black oil, modified black oil (MBO), and compositional models. These models are used in calculations that allow engineers to optimize and forecast well and reservoir performance (e.g., reservoir simulation calculations, material balance, nodal analysis, surface facilities, etc.). The choice of which model is dependent on fluid type and the production process (e.g., depletion, water injection, gas injection, etc.). Based on close to 2,000 reservoir fluid samples collected from different basins and locations, this paper presents some conclusions on the occurrence of reservoir fluids. It also reviews the common methods used to classify reservoir fluid types. Based on new criteria related to the production behavior of different fluids and economic considerations, an updated classification of reservoir fluid types is presented in the paper. Recommendations on the use of different PVT models to simulate the behavior of different reservoir fluid types are discussed. Each PVT model requirement is highlighted. Available methods for the calculation of PVT properties from each model are also discussed. Practical recommendations and tips on how to control the calculations to achieve the most accurate results are given.

Keywords: PVT models, fluid types, PVT properties, fluids classification

Procedia PDF Downloads 60

Authors: Yanchun Zuo, Yingao Liu, Wei Liu, Le Yu, Run Huang, Lixin Guo

Abstract:

Electric field detection is important in many application scenarios. The traditional strategy is measuring the electric field with a man walking around in the area under test. This strategy cannot provide a satisfactory measurement accuracy. To solve the mentioned problem, an autopilot measurement system is divided. A mini-car is produced, which can travel in the area under test according to respect to the program within the CPU. The electric field measurement platform (EFMP) carries a central computer, two horn antennas, and a vector network analyzer. The mini-car stop at the sampling points according to the preset. When the car stops, the EFMP probes the electric field and stores data on the hard disk. After all the sampling points are traversed, an electric field map can be plotted. The proposed system can give an accurate field distribution description of the chamber.

Keywords: autopilot mini-car measurement system, electric field detection, field map, static zone measurement

Procedia PDF Downloads 92

Authors: Faiza Belbachir, Thibault Schienhinski

Abstract:

The massive development of online social networks allows users to post and share their opinions on various topics. With this huge volume of opinion, it is interesting to extract and interpret these information for different domains, e.g., product and service benchmarking, politic, system of recommendation. This is why opinion detection is one of the most important research tasks. It consists on differentiating between opinion data and factual data. The difficulty of this task is to determine an approach which returns opinionated document. Generally, there are two approaches used for opinion detection i.e. Lexical based approaches and Machine Learning based approaches. In Lexical based approaches, a dictionary of sentimental words is used, words are associated with weights. The opinion score of document is derived by the occurrence of words from this dictionary. In Machine learning approaches, usually a classifier is trained using a set of annotated document containing sentiment, and features such as n-grams of words, part-of-speech tags, and logical forms. Majority of these works are based on documents text to determine opinion score but dont take into account if these texts are really correct. Thus, it is interesting to exploit other information to improve opinion detection. In our work, we will develop a new way to consider the opinion score. We introduce the notion of trust score. We determine opinionated documents but also if these opinions are really trustable information in relation with topics. For that we use lexical SentiWordNet to calculate opinion and trust scores, we compute different features about users like (numbers of their comments, numbers of their useful comments, Average useful review). After that, we combine opinion score and trust score to obtain a final score. We applied our method to detect trust opinions in TRIPADVISOR collection. Our experimental results report that the combination between opinion score and trust score improves opinion detection.

Keywords: Tripadvisor, opinion detection, SentiWordNet, trust score

Procedia PDF Downloads 187

Authors: Asma A. Alsufyani, Maram A. Alharthi, Maha J. Althobaiti, Manal S. Alharthi, Huda Rizq

Abstract:

Twitter is one of the most popular microblogging sites that allows users to publish short text messages called 'tweets'. Increasing the number of accounts to follow (followings) increases the number of tweets that will be displayed from different topics in an unclassified manner in the timeline of the user. Therefore, it can be a vital solution for many Twitter users to have their tweets in a timeline classified into general categories to save the user’s time and to provide easy and quick access to tweets based on topics. In this paper, we developed a classifier for timeline tweets trained on a dataset consisting of 3600 tweets in total, which were collected from Saudi Twitter and annotated manually. We experimented with the well-known Bag-of-Words approach to text classification, and we used support vector machines (SVM) in the training process. The trained classifier performed well on a test dataset, with an average F1-measure equal to 92.3%. The classifier has been integrated into an application, which practically proved the classifier’s ability to classify timeline tweets of the user.

Keywords: corpus creation, feature extraction, machine learning, short text classification, social media, support vector machine, Twitter

Procedia PDF Downloads 140

Authors: E. M. Rifqi Wilda Pradana, Bagus Bayu Prabowo, Meida Riski Pujiyati, Efraim Maykhel Hagana Ginting, Virgiawan Arya Hangga Reksa

Abstract:

The VLF (Very Low Frequency) method is an electromagnetic method that uses low frequencies between 10-30 KHz which results in a fairly deep penetration. In this study, the VLF method was used for zonation of disaster-prone areas by identifying geological structures in the form of faults. Data acquisition was carried out in Trimulyo Region, Jetis District, Bantul Regency, Special Region of Yogyakarta, Indonesia with 8 measurement paths. This study uses wave transmitters from Japan and Australia to obtain Tilt and Elipt values that can be used to create RAE (Rapat Arus Ekuivalen or Current Density) sections that can be used to identify areas that are easily crossed by electric current. This section will indicate the existence of a geological structure in the form of faults in the study area which is characterized by a high RAE value. In data processing of VLF method, it is obtained Tilt vs Elliptical graph and Moving Average (MA) Tilt vs Moving Average (MA) Elipt graph of each path that shows a fluctuating pattern and does not show any intersection at all. Data processing uses Matlab software and obtained areas with low RAE values that are 0%-6% which shows medium with low conductivity and high resistivity and can be interpreted as sandstone, claystone, and tuff lithology which is part of the Semilir Formation. Whereas a high RAE value of 10% -16% which shows a medium with high conductivity and low resistivity can be interpreted as a fault zone filled with fluid. The existence of the fault zone is strengthened by the discovery of a normal fault on the surface with strike N550W and dip 630E at coordinates X= 433256 and Y= 9127722 so that the activities of residents in the zone such as housing, mining activities and other activities can be avoided to reduce the risk of natural disasters.

Keywords: current density, faults, very low frequency, zonation

Procedia PDF Downloads 161

Authors: Koray U. Erbas

Abstract:

Neural networks are appealing for many applications since they are able to learn complex non-linear relationships between input and output data. As the number of neurons and layers in a neural network increase, it is possible to represent more complex relationships with automatically extracted features. Nowadays Deep Neural Networks (DNNs) are widely used in Computer Vision problems such as; classification, object detection, segmentation image editing etc. In this work, Facial Emotion Recognition task is performed by proposed Convolutional Neural Network (CNN)-based DNN architecture using FER2013 Dataset. Moreover, the effects of different hyperparameters (activation function, kernel size, initializer, batch size and network size) are investigated and ablation study results for Pooling Layer, Dropout and Batch Normalization are presented.

Keywords: convolutional neural network, deep learning, deep learning based FER, facial emotion recognition

Procedia PDF Downloads 248

Authors: Radhia Toujani, Jalel Akaichi

Abstract:

Social Networks generally present a hierarchy of communities. To determine these communities and the relationship between them, detection algorithms should be applied. Most of the existing algorithms, proposed for hierarchical communities identification, are based on either agglomerative clustering or divisive clustering. In this paper, we present a hybrid hierarchical clustering approach for community detection based on both bottom-up and bottom-down clustering. Obviously, our approach provides more relevant community structure than hierarchical method which considers only divisive or agglomerative clustering to identify communities. Moreover, we performed some comparative experiments to enhance the quality of the clustering results and to show the effectiveness of our algorithm.

Keywords: agglomerative hierarchical clustering, community structure, divisive hierarchical clustering, hybrid hierarchical clustering, opinion mining, social network, social network analysis

Procedia PDF Downloads 350

Authors: Yanxue Shang, Jingbin Zeng

Abstract:

Chlorobenzene homologues (CBHs) are a category of environmental pollutants that can not be ignored. They can stay in the environment for a long period and are potentially carcinogenic. The traditional degradation method of CBHs is dechlorination followed by sample preparation and analysis. This is not only time-consuming and laborious, but the detection and analysis processes are used in conjunction with large-scale instruments. Therefore, this can not achieve rapid and low-cost detection. Compared with traditional sensing methods, colorimetric sensing is simpler and more convenient. In recent years, chromaticity sensors based on fluorescence have attracted more and more attention. Compared with sensing methods based on changes in fluorescence intensity, changes in color gradients are easier to recognize by the naked eye. Accordingly, this work proposes to use single drop microextraction (SDME) technology to solve the above problems. After the dechlorination reaction was completed, the organic droplet extracts Cl⁻ and realizes fluorescence colorimetric sensing at the same time. This method was integrated sample processing and visual in-situ detection, simplifying the detection process. As a fluorescence colorimetric sensor material, CsPbBr₃ was encapsulated in MOF-5 to construct CsPbBr₃@MOF-5 fluorescence colorimetric composite. Then the fluorescence colorimetric sensor was constructed by dispersing the composite in SDME organic droplets. When the Br⁻ in CsPbBr₃ exchanges with Cl⁻ produced by the dechlorination reactions, it is converted into CsPbCl₃. The fluorescence color of the single droplet of SDME will change from green to blue emission, thereby realizing visual observation. Therein, SDME can enhance the concentration and enrichment of Cl⁻ and instead of sample pretreatment. The fluorescence color change of CsPbBr₃@MOF-5 can replace the detection process of large-scale instruments to achieve real-time rapid detection. Due to the absorption ability of MOF-5, it can not only improve the stability of CsPbBr₃, but induce the adsorption of Cl⁻. Simultaneously, accelerate the exchange of Br- and Cl⁻ in CsPbBr₃ and the detection process of Cl⁻. The absorption process was verified by density functional theory (DFT) calculations. This method exhibits exceptional linearity for Cl⁻ in the range of 10⁻² - 10⁻⁶ M (10000 μM - 1 μM) with a limit of detection of 10⁻⁷ M. Whereafter, the dechlorination reactions of different kinds of CBHs were also carried out with this method, and all had satisfactory detection ability. Also verified the accuracy by gas chromatography (GC), and it was found that the SDME we developed in this work had high credibility. In summary, the in-situ visualization method of dechlorination reaction detection was a combination of sample processing and fluorescence colorimetric sensing. Thus, the strategy researched herein represents a promising method for the visual detection of dechlorination reactions and can be extended for applications in environments, chemical industries, and foods.

Keywords: chlorobenzene homologues, colorimetric sensor, metal halide perovskite, metal-organic frameworks, single drop microextraction

Procedia PDF Downloads 133

Authors: Atish Bagchi, Siva Chandrasekaran

Abstract:

Background: According to World Bank national accounts data, the estimated global manufacturing value-added output in 2020 was 13.74 trillion USD. These manufacturing processes are monitored, modelled, and controlled by advanced, real-time, computer-based systems, e.g., Industrial IoT, PLC, SCADA, etc. These systems measure and manipulate a set of physical variables, e.g., temperature, pressure, etc. Despite the use of IoT, SCADA etc., in manufacturing, studies suggest that unplanned downtime leads to economic losses of approximately 864 billion USD each year. Therefore, real-time, accurate detection, classification and prediction of machine behaviour are needed to minimise financial losses. Although vast literature exists on time-series data processing using machine learning, the challenges faced by the industries that lead to unplanned downtimes are: The current algorithms do not efficiently handle the high-volume streaming data from industrial IoTsensors and were tested on static and simulated datasets. While the existing algorithms can detect significant 'point' outliers, most do not handle contextual outliers (e.g., values within normal range but happening at an unexpected time of day) or subtle changes in machine behaviour. Machines are revamped periodically as part of planned maintenance programmes, which change the assumptions on which original AI models were created and trained. Aim: This research study aims to deliver a Graph Neural Network(GNN)based hybrid forecasting model that interfaces with the real-time machine control systemand can detect, predict machine behaviour and behavioural changes (anomalies) in real-time. This research will help manufacturing industries and utilities, e.g., water, electricity etc., reduce unplanned downtimes and consequential financial losses. Method: The data stored within a process control system, e.g., Industrial-IoT, Data Historian, is generally sampled during data acquisition from the sensor (source) and whenpersistingin the Data Historian to optimise storage and query performance. The sampling may inadvertently discard values that might contain subtle aspects of behavioural changes in machines. This research proposed a hybrid forecasting and classification model which combines the expressive and extrapolation capability of GNN enhanced with the estimates of entropy and spectral changes in the sampled data and additional temporal contexts to reconstruct the likely temporal trajectory of machine behavioural changes. The proposed real-time model belongs to the Deep Learning category of machine learning and interfaces with the sensors directly or through 'Process Data Historian', SCADA etc., to perform forecasting and classification tasks. Results: The model was interfaced with a Data Historianholding time-series data from 4flow sensors within a water treatment plantfor45 days. The recorded sampling interval for a sensor varied from 10 sec to 30 min. Approximately 65% of the available data was used for training the model, 20% for validation, and the rest for testing. The model identified the anomalies within the water treatment plant and predicted the plant's performance. These results were compared with the data reported by the plant SCADA-Historian system and the official data reported by the plant authorities. The model's accuracy was much higher (20%) than that reported by the SCADA-Historian system and matched the validated results declared by the plant auditors. Conclusions: The research demonstrates that a hybrid GNN based approach enhanced with entropy calculation and spectral information can effectively detect and predict a machine's behavioural changes. The model can interface with a plant's 'process control system' in real-time to perform forecasting and classification tasks to aid the asset management engineers to operate their machines more efficiently and reduce unplanned downtimes. A series of trialsare planned for this model in the future in other manufacturing industries.

Keywords: GNN, Entropy, anomaly detection, industrial time-series, AI, IoT, Industry 4.0, Machine Learning

Procedia PDF Downloads 135

Authors: Seyhan Karaçavuş, Bülent Yılmaz, Ömer Kayaaltı, Semra İçer, Arzu Taşdemir, Oğuzhan Ayyıldız, Kübra Eset, Eser Kaya

Abstract:

In this study, our goal was to perform tumor staging and subtype determination automatically using different texture analysis approaches for a very common cancer type, i.e., non-small cell lung carcinoma (NSCLC). Especially, we introduced a texture analysis approach, called Law’s texture filter, to be used in this context for the first time. The 18F-FDG PET images of 42 patients with NSCLC were evaluated. The number of patients for each tumor stage, i.e., I-II, III or IV, was 14. The patients had ~45% adenocarcinoma (ADC) and ~55% squamous cell carcinoma (SqCCs). MATLAB technical computing language was employed in the extraction of 51 features by using first order statistics (FOS), gray-level co-occurrence matrix (GLCM), gray-level run-length matrix (GLRLM), and Laws’ texture filters. The feature selection method employed was the sequential forward selection (SFS). Selected textural features were used in the automatic classification by k-nearest neighbors (k-NN) and support vector machines (SVM). In the automatic classification of tumor stage, the accuracy was approximately 59.5% with k-NN classifier (k=3) and 69% with SVM (with one versus one paradigm), using 5 features. In the automatic classification of tumor subtype, the accuracy was around 92.7% with SVM one vs. one. Texture analysis of FDG-PET images might be used, in addition to metabolic parameters as an objective tool to assess tumor histopathological characteristics and in automatic classification of tumor stage and subtype.

Keywords: cancer stage, cancer cell type, non-small cell lung carcinoma, PET, texture analysis

Procedia PDF Downloads 314

Authors: Suveen Kumar

Abstract:

Oral cancer (OC) is the sixth most death causing cancer in world which includes tumour of lips, floor of the mouth, tongue, palate, cheeks, sinuses, throat, etc. Conventionally, the techniques used for OC detection are toluidine blue staining, biopsy, liquid-based cytology, visual attachments, etc., however these are limited by their highly invasive nature, low sensitivity, time consumption, sophisticated instrument handling, sample processing and high cost. Therefore, we developed biosensing chips for non-invasive detection of OC via CYFRA-21-1 biomarker. CYFRA-21-1 (molecular weight: 40 kDa) is secreted in saliva of OC patients which is a non-invasive biological fluid with a cut-off value of 3.8 ng mL-1, above which the subjects will be suffering from oral cancer. Therefore, in first work, 3-aminopropyl triethoxy silane (APTES) functionalized zirconia (ZrO2) nanoparticles (APTES/nZrO2) were used to successfully detect CYFRA-21-1 in a linear detection range (LDR) of 2-16 ng mL-1 with sensitivity of 2.2 µA mL ng-1. Successively, APTES/nZrO2-RGO was employed to prevent agglomeration of ZrO2 by providing high surface area reduced graphene oxide (RGO) support and much wider LDR (2-22 ng mL-1) was obtained with remarkable limit of detection (LOD) as 0.12 ng mL-1. Further, APTES/nY2O3/ITO platform was used for oral cancer bioseneor development. The developed biosensor (BSA/anti-CYFRA-21-1/APTES/nY2O3/ITO) have wider LDR (0.01-50 ng mL-1) with remarkable limit of detection (LOD) as 0.01 ng mL-1. To improve the sensitivity of the biosensing platform, nanocomposite of yattria stabilized nanostructured zirconia-reduced graphene oxide (nYZR) based biosensor has been developed. The developed biosensing chip having ability to detect CYFRA-21-1 biomolecules in the range of 0.01-50 ng mL-1, LOD of 7.2 pg mL-1 with sensitivity of 200 µA mL ng-1. Further, the applicability of the fabricated biosensing chips were also checked through real sample (saliva) analysis of OC patients and the obtained results showed good correlation with the standard protein detection enzyme linked immunosorbent assay (ELISA) technique.

Keywords: non-invasive, oral cancer, nanomaterials, biosensor, biochip

Procedia PDF Downloads 116

Authors: Michael Zimba

Abstract:

A duplicated image region may be subjected to a number of attacks such as noise addition, compression, reflection, rotation, and scaling with the intention of either merely mating it to its targeted neighborhood or preventing its detection. In this paper, we present an effective and robust method of detecting duplicated regions inclusive of those affected by the various attacks. In order to reduce the dimension of the image, the proposed algorithm firstly performs discrete wavelet transform, DWT, of a suspicious image. However, unlike most existing copy move image forgery (CMIF) detection algorithms operating in the DWT domain which extract only the low frequency sub-band of the DWT of the suspicious image thereby leaving valuable information in the other three sub-bands, the proposed algorithm simultaneously extracts features from all the four sub-bands. The extracted features are not only more accurate representation of image regions but also robust to additive noise, JPEG compression, and affine transformation. Furthermore, principal component analysis-eigenvalue decomposition, PCA-EVD, is applied to reduce the dimension of the features. The extracted features are then sorted using the more computationally efficient Radix Sort algorithm. Finally, same affine transformation selection, SATS, a duplication verification method, is applied to detect duplicated regions. The proposed algorithm is not only fast but also more robust to attacks compared to the related CMIF detection algorithms. The experimental results show high detection rates.

Keywords: affine transformation, discrete wavelet transform, radix sort, SATS

Procedia PDF Downloads 220

Authors: Temesgen Geremew

Abstract:

This study investigates the evolution of polythiophene nanoparticles (PThNPs) as surface-enhanced Raman spectroscopy (SERS) sensors for Pb²⁺ detection in river water. We analyze the PThNPs' optical, structural, and morphological properties at different stages of their development to understand their SERS performance. Techniques like UV-Vis spectroscopy, Fourier-transform infrared spectroscopy (FTIR), X-ray diffraction (XRD), and scanning electron microscopy (SEM) are employed for characterization. The SERS sensitivity towards Pb²⁺ is evaluated by monitoring the peak intensity of a specific Raman band upon increasing metal ion concentration. The study aims to elucidate the relationship between the PThNPs' characteristics and their SERS efficiency for Pb²⁺ detection, paving the way for optimizing their design and fabrication for improved sensing performance in real-world environmental monitoring applications.

Keywords: polythiophene, Pb2+, SERS, nanoparticles

Procedia PDF Downloads 38

Authors: Eiman Kattan, Hong Wei

Abstract:

In using Convolutional Neural Network (CNN) for classification, there is a set of hyperparameters available for the configuration purpose. This study aims to evaluate the impact of a range of parameters in CNN architecture i.e. AlexNet on land cover classification based on four remotely sensed datasets. The evaluation tests the influence of a set of hyperparameters on the classification performance. The parameters concerned are epoch values, batch size, and convolutional filter size against input image size. Thus, a set of experiments were conducted to specify the effectiveness of the selected parameters using two implementing approaches, named pertained and fine-tuned. We first explore the number of epochs under several selected batch size values (32, 64, 128 and 200). The impact of kernel size of convolutional filters (1, 3, 5, 7, 10, 15, 20, 25 and 30) was evaluated against the image size under testing (64, 96, 128, 180 and 224), which gave us insight of the relationship between the size of convolutional filters and image size. To generalise the validation, four remote sensing datasets, AID, RSD, UCMerced and RSCCN, which have different land covers and are publicly available, were used in the experiments. These datasets have a wide diversity of input data, such as number of classes, amount of labelled data, and texture patterns. A specifically designed interactive deep learning GPU training platform for image classification (Nvidia Digit) was employed in the experiments. It has shown efficiency in both training and testing. The results have shown that increasing the number of epochs leads to a higher accuracy rate, as expected. However, the convergence state is highly related to datasets. For the batch size evaluation, it has shown that a larger batch size slightly decreases the classification accuracy compared to a small batch size. For example, selecting the value 32 as the batch size on the RSCCN dataset achieves the accuracy rate of 90.34 % at the 11th epoch while decreasing the epoch value to one makes the accuracy rate drop to 74%. On the other extreme, setting an increased value of batch size to 200 decreases the accuracy rate at the 11th epoch is 86.5%, and 63% when using one epoch only. On the other hand, selecting the kernel size is loosely related to data set. From a practical point of view, the filter size 20 produces 70.4286%. The last performed image size experiment shows a dependency in the accuracy improvement. However, an expensive performance gain had been noticed. The represented conclusion opens the opportunities toward a better classification performance in various applications such as planetary remote sensing.

Keywords: CNNs, hyperparamters, remote sensing, land cover, land use

Procedia PDF Downloads 159

Authors: Daliyah S. Aljutaili, Redna A. Almutlaq, Suha A. Alharbi, Dina M. Ibrahim

Abstract:

All currencies around the world look very different from each other. For instance, the size, color, and pattern of the paper are different. With the development of modern banking services, automatic methods for paper currency recognition become important in many applications like vending machines. One of the currency recognition architecture’s phases is Feature detection and description. There are many algorithms that are used for this phase, but they still have some disadvantages. This paper proposes a feature detection algorithm, which merges the advantages given in the current SIFT and SURF algorithms, which we call, Speeded up Robust Scale-Invariant Feature Transform (SR-SIFT) algorithm. Our proposed SR-SIFT algorithm overcomes the problems of both the SIFT and SURF algorithms. The proposed algorithm aims to speed up the SIFT feature detection algorithm and keep it robust. Simulation results demonstrate that the proposed SR-SIFT algorithm decreases the average response time, especially in small and minimum number of best key points, increases the distribution of the number of best key points on the surface of the currency. Furthermore, the proposed algorithm increases the accuracy of the true best point distribution inside the currency edge than the other two algorithms.

Keywords: currency recognition, feature detection and description, SIFT algorithm, SURF algorithm, speeded up and robust features

Procedia PDF Downloads 225

Authors: I. Stathopoulos, V. Syrgiamiotis, E. Karavasilis, A. Ploussi, I. Nikas, C. Hatzigiorgi, K. Platoni, E. P. Efstathopoulos

Abstract:

Introduction: Brain and central nervous system (CNS) tumors form the second most common group of cancer in children, accounting for 30% of all childhood cancers. MRI is the key imaging technique used for the visualization and management of pediatric brain tumors. Initial characterization of tumors from MRI scans is usually performed via a radiologist’s visual assessment. However, different brain tumor types do not always demonstrate clear differences in visual appearance. Using only conventional MRI to provide a definite diagnosis could potentially lead to inaccurate results, and so histopathological examination of biopsy samples is currently considered to be the gold standard for obtaining definite diagnoses. Machine learning is defined as the study of computational algorithms that can use, complex or not, mathematical relationships and patterns from empirical and scientific data to make reliable decisions. Concerning the above, machine learning techniques could provide effective and accurate ways to automate and speed up the analysis and diagnosis for medical images. Machine learning applications in radiology are or could potentially be useful in practice for medical image segmentation and registration, computer-aided detection and diagnosis systems for CT, MR or radiography images and functional MR (fMRI) images for brain activity analysis and neurological disease diagnosis. Purpose: The objective of this study is to provide an automated tool, which may assist in the imaging evaluation and classification of brain neoplasms in pediatric patients by determining the glioma type, grade and differentiating between different brain tissue types. Moreover, a future purpose is to present an alternative way of quick and accurate diagnosis in order to save time and resources in the daily medical workflow. Materials and Methods: A cohort, of 80 pediatric patients with a diagnosis of posterior fossa tumor, was used: 20 ependymomas, 20 astrocytomas, 20 medulloblastomas and 20 healthy children. The MR sequences used, for every single patient, were the following: axial T1-weighted (T1), axial T2-weighted (T2), FluidAttenuated Inversion Recovery (FLAIR), axial diffusion weighted images (DWI), axial contrast-enhanced T1-weighted (T1ce). From every sequence only a principal slice was used that manually traced by two expert radiologists. Image acquisition was carried out on a GE HDxt 1.5-T scanner. The images were preprocessed following a number of steps including noise reduction, bias-field correction, thresholding, coregistration of all sequences (T1, T2, T1ce, FLAIR, DWI), skull stripping, and histogram matching. A large number of features for investigation were chosen, which included age, tumor shape characteristics, image intensity characteristics and texture features. After selecting the features for achieving the highest accuracy using the least number of variables, four machine learning classification algorithms were used: k-Nearest Neighbour, Support-Vector Machines, C4.5 Decision Tree and Convolutional Neural Network. The machine learning schemes and the image analysis are implemented in the WEKA platform and MatLab platform respectively. Results-Conclusions: The results and the accuracy of images classification for each type of glioma by the four different algorithms are still on process.

Keywords: image classification, machine learning algorithms, pediatric MRI, pediatric oncology

Procedia PDF Downloads 139

Authors: Jianuo Sun, Jinghan Wang, Sirui Zhang, Chenhan Xu, Hongxia Hao, Hong Zhou

Abstract:

In recent years, the diversification and industrialization of drug-related crimes have posed significant threats to public health and safety globally. The widespread and increasingly younger demographics of drug users and the persistence of drug-impaired driving incidents underscore the urgency of this issue. Drug detection, a specialized forensic activity, is pivotal in identifying and analyzing substances involved in drug crimes. It relies on pharmacological and chemical knowledge and employs analytical chemistry and modern detection techniques. However, current drug detection methods are limited by their inability to perform semi-quantitative, real-time field analyses. They require extensive, complex laboratory-based preprocessing, expensive equipment, and specialized personnel and are hindered by long processing times. This study introduces an alternative approach using nucleic acid aptamers and Aggregation-Induced Emission (AIE) technology. Nucleic acid aptamers, selected artificially for their specific binding to target molecules and stable spatial structures, represent a new generation of biosensors following antibodies. Rapid advancements in AIE technology, particularly in tetraphenyl ethene-based luminous, offer simplicity in synthesis and versatility in modifications, making them ideal for fluorescence analysis. This work successfully synthesized, isolated, and purified an AIE molecule and constructed a probe comprising the AIE molecule, nucleic acid aptamers, and exonuclease for cocaine detection. The probe demonstrated significant relative fluorescence intensity changes and selectivity towards cocaine over other drugs. Using 4-Butoxytriethylammonium Bromide Tetraphenylethene (TPE-TTA) as the fluorescent probe, the aptamer as the recognition unit, and Exo I as an auxiliary, the system achieved rapid detection of cocaine within 5 mins in aqueous and urine, with detection limits of 1.0 and 5.0 µmol/L respectively. The probe-maintained stability and interference resistance in urine, enabling quantitative cocaine detection within a certain concentration range. This fluorescent sensor significantly reduces sample preprocessing time, offers a basis for rapid onsite cocaine detection, and promises potential for miniaturized testing setups.

Keywords: drug detection, aggregation-induced emission (AIE), nucleic acid aptamer, exonuclease, cocaine

Procedia PDF Downloads 51

Authors: Osama Mohamed Elrajubi, Idris El-Feghi, Mohamed Abu Baker Saghayer

Abstract:

This paper describes an identification of specific shapes within binary images using the morphological Hit-or-Miss Transform (HMT). Hit-or-Miss transform is a general binary morphological operation that can be used in searching of particular patterns of foreground and background pixels in an image. It is actually a basic operation of binary morphology since almost all other binary morphological operators are derived from it. The input of this method is a binary image and a structuring element (a template which will be searched in a binary image) while the output is another binary image. In this paper a modification of Hit-or-Miss transform has been proposed. The accuracy of algorithm is adjusted according to the similarity of the template and the sought template. The implementation of this method has been done by C language. The algorithm has been tested on several images and the results have shown that this new method can be used for similar shape detection.

Keywords: hit-or-miss operator transform, HMT, binary morphological operation, shape detection, binary images processing

Procedia PDF Downloads 320

Authors: Nimet Yildirim Tirgil, Ahmed Busnaina, April Z. Gu

Abstract:

Environmental waters are monitored worldwide to protect people from infectious diseases primarily caused by enteric pathogens. All long, Escherichia coli (E. coli) is a good indicator for potential enteric pathogens in waters. Thus, a rapid and simple detection method for E. coli is very important to predict the pathogen contamination. In this study, to the best of our knowledge, as the first time we developed a rapid, direct and reusable SWCNTs (single walled carbon nanotubes) based biosensor system for sensitive and selective E. coli detection in water samples. We use a novel and newly developed flexible biosensor device which was fabricated by high-rate nanoscale offset printing process using directed assembly and transfer of SWCNTs. By simple directed assembly and non-covalent functionalization, aptamer (biorecognition element that specifically distinguish the E. coli O157:H7 strain from other pathogens) based SWCNTs biosensor system was designed and was further evaluated for environmental applications with simple and cost-effective steps. The two gold electrode terminals and SWCNTs-bridge between them allow continuous resistance response monitoring for the E. coli detection. The detection procedure is based on competitive mode detection. A known concentration of aptamer and E. coli cells were mixed and after a certain time filtered. The rest of free aptamers injected to the system. With hybridization of the free aptamers and their SWCNTs surface immobilized probe DNA (complementary-DNA for E. coli aptamer), we can monitor the resistance difference which is proportional to the amount of the E. coli. Thus, we can detect the E. coli without injecting it directly onto the sensing surface, and we could protect the electrode surface from the aggregation of target bacteria or other pollutants that may come from real wastewater samples. After optimization experiments, the linear detection range was determined from 2 cfu/ml to 10⁵ cfu/ml with higher than 0.98 R² value. The system was regenerated successfully with 5 % SDS solution over 100 times without any significant deterioration of the sensor performance. The developed system had high specificity towards E. coli (less than 20 % signal with other pathogens), and it could be applied to real water samples with 86 to 101 % recovery and 3 to 18 % cv values (n=3).

Keywords: aptamer, E. coli, environmental detection, nanobiosensor, SWCTs

Procedia PDF Downloads 181

Authors: Ebenezer Obaloluwa Olaniyi

Abstract:

In this paper, an intelligent system has been designed to grade apple based on either its defective or healthy for production in food processing. This paper is segmented into two different phase. In the first phase, the image processing techniques were employed to extract the necessary features required in the apple. These techniques include grayscale conversion, segmentation where a threshold value is chosen to separate the foreground of the images from the background. Then edge detection was also employed to bring out the features in the images. These extracted features were then fed into the neural network in the second phase of the paper. The second phase is a classification phase where neural network employed to classify the defective apple from the healthy apple. In this phase, the network was trained with back propagation and tested with feed forward network. The recognition rate obtained from our system shows that our system is more accurate and faster as compared with previous work.

Keywords: image processing, neural network, apple, intelligent system

Procedia PDF Downloads 388

Authors: Mohamad Syahrizal Ahmad, Illyas M. Isa

Abstract:

In this paper, a multiwall carbon nanotubes (MWCNT) paste electrode modified by a Zn layered hydroxide-ferulate (ZLH-F) was used for detection of hydroquinone (HQ). The morphology and characteristic of the ZLH-F/MWCNT were investigated by scanning electron microscope (SEM), transmission electron microscope (TEM) and square wave voltammetry (SWV). Under optimal conditions, the SWV response showed linear plot for HQ concentration in the range of 1.0×10⁻⁵ M – 1.0×10⁻³ M. The detection limit was found to be 5.7×10⁻⁶ M and correlation coefficient of 0.9957. The glucose, fructose, sucrose, bisphenol A, acetaminophen, lysine, NO₃⁻, Cl⁻ and SO₄²⁻ did not interfere the HQ response. This modified electrode can be used to determine HQ content in wastewater and cosmetic cream with range of recovery 97.8% - 103.0%.

Keywords: 1, 4-dihydroxybenzene, hydroquinone, multiwall carbon nanotubes, square wave voltammetry

Procedia PDF Downloads 218

Authors: Yakin Hajlaoui, Richard Labib, Jean-François Plante, Michel Gamache

Abstract:

This study introduces the Multi-Layer Inverse Distance Weighting Model (ML-IDW), inspired by the mathematical formulation of both multi-layer neural networks (ML-NNs) and Inverse Distance Weighting model (IDW). ML-IDW leverages ML-NNs' processing capabilities, characterized by compositions of learnable non-linear functions applied to input features, and incorporates IDW's ability to learn anisotropic spatial dependencies, presenting a promising solution for nonlinear spatial interpolation and learning from complex spatial data. it employ gradient descent and backpropagation to train ML-IDW, comparing its performance against conventional spatial interpolation models such as Kriging and standard IDW on regression and classification tasks using simulated spatial datasets of varying complexity. the results highlight the efficacy of ML-IDW, particularly in handling complex spatial datasets, exhibiting lower mean square error in regression and higher F1 score in classification.

Keywords: deep learning, multi-layer neural networks, gradient descent, spatial interpolation, inverse distance weighting

Procedia PDF Downloads 32

Authors: Jamileh Vasheghani Farahani

Abstract:

The main purpose of this research is to show the current active faults and active tectonic of the area by three seismic networks in Tehran region: 1-Tehran Disaster Mitigation and Management Organization (TDMMO), 2-Broadband Iranian National Seismic Network Center (BIN), 3-Iranian Seismological Center (IRSC). In this study, we analyzed microearthquakes happened in Tehran city and its surroundings using the Tehran networks from 1996 to 2015. We found some active faults and trends in the region. There is a 200-year history of historical earthquakes in Tehran. Historical and instrumental seismicity show that the east of Tehran is more active than the west. The Mosha fault in the North of Tehran is one of the active faults of the central Alborz. Moreover, other major faults in the region are Kahrizak, Eyvanakey, Parchin and North Tehran faults. An important seismicity region is an intersection of the Mosha and North Tehran fault systems (Kalan village in Lavasan). This region shows a cluster of microearthquakes. According to the historical and microseismic events analyzed in this research, there is a seismic gap in SE of Tehran. The empirical relationship is used to assess the Mmax based on the rupture length. There is a probability of occurrence of a strong motion of 7.0 to 7.5 magnitudes in the region (based on the assessed capability of the major faults such as Parchin and Eyvanekey faults and historical earthquakes).

Keywords: Iran, major faults, microseismicity, Tehran

Procedia PDF Downloads 358

Authors: Sevilay Çankaya

Abstract:

The detection of deception in children’s interviews is essential for statement veracity. The widely used method for deception detection is building cognitive load, which is the logic of the cognitive interview (CI), and its effectiveness for adults is approved. This meta-analysis delves into the effectiveness of inducing cognitive load as a means of enhancing veracity detection during interviews with children. Additionally, the effectiveness of cognitive load on children's total number of events recalled is assessed as a second part of the analysis. The current meta-analysis includes ten effect sizes from search using databases. For the effect size calculation, Hedge’s g was used with a random effect model by using CMA version 2. Heterogeneity analysis was conducted to detect potential moderators. The overall result indicated that cognitive load had no significant effect on veracity outcomes (g =0.052, 95% CI [-.006,1.25]). However, a high level of heterogeneity was found (I² = 92%). Age, participants’ characteristics, interview setting, and characteristics of the interviewer were coded as possible moderators to explain variance. Age was significant moderator (β = .021; p = .03, R2 = 75%) but the analysis did not reveal statistically significant effects for other potential moderators: participants’ characteristics (Q = 0.106, df = 1, p = .744), interview setting (Q = 2.04, df = 1, p = .154), and characteristics of interviewer (Q = 2.96, df = 1, p = .086). For the second outcome, the total number of events recalled, the overall effect was significant (g =4.121, 95% CI [2.256,5.985]). The cognitive load was effective in total recalled events when interviewing with children. All in all, while age plays a crucial role in determining the impact of cognitive load on veracity, the surrounding context, interviewer attributes, and inherent participant traits may not significantly alter the relationship. These findings throw light on the need for more focused, age-specific methods when using cognitive load measures. It may be possible to improve the precision and dependability of deceit detection in children's interviews with the help of more studies in this field.

Keywords: deceit detection, cognitive load, memory recall, children interviews, meta-analysis

Procedia PDF Downloads 47

Authors: Josef Steinberger, Vanni Zavarella, Hristo Tanev

Abstract:

Natural disasters cause tens of thousands of victims and massive material damages. We refer to all those events caused by natural disasters, such as damage on people, infrastructure, vehicles, services and resource supply, as micro events. This paper addresses the problem of micro - event detection in online media sources. We present a natural language grammar learning algorithm and apply it to online news. The algorithm in question is based on distributional clustering and detection of word collocations. We also explore the extraction of micro-events from social media and describe a Twitter mining robot, who uses combinations of keywords to detect tweets which talk about effects of disasters.

Keywords: online news, natural language processing, machine learning, event extraction, crisis computing, disaster effects, Twitter

Procedia PDF Downloads 472

Authors: Tamas Madl

Abstract:

Sudden cardiac death (SCD) constitutes a large proportion of cardiovascular mortalities, provides little advance warning, and the risk is difficult to recognize based on ubiquitous, low cost medical equipment such as the standard, 12-lead, ten second ECG. Autonomic abnormalities have been shown to be strongly predictive of SCD risk; yet current methods are not trivially applicable to the brevity and low temporal and electrical resolution of standard ECGs. Here, we build horizontal visibility graph representations of very short inter-beat interval time series, and perform unsuper- vised representation learning in order to convert these variable size objects into fixed-length vectors preserving similarity rela- tions. We show that such representations facilitate classification into healthy vs. at-risk patients on two different datasets, the Mul- tiparameter Intelligent Monitoring in Intensive Care II and the PhysioNet Sudden Cardiac Death Holter Database. Our results suggest that graph representation learning of heartbeat interval time series facilitates robust classification even in sequences as short as ten seconds.

Keywords: sudden cardiac death, heart rate variability, ECG analysis, time series classification

Procedia PDF Downloads 225

Authors: Pam Bitrus Marcus

Abstract:

Compounds in Berom, a Niger-Congo language that is spoken in parts of central Nigeria, have been understudied, and the semantics of N-V nominal compounds have not been sufficiently delineated. This study describes the lexical classification of compounds in Berom and, specifically, examines the semantics of nominal compounds with N-V constituents. The study relied on a data set of 200 compounds that were drawn from Bere Naha (a newsletter publication in Berom). Contrary to the nominalization process in defining the lexical class of compounds in languages, the study revealed that verbal and adjectival classes of compounds are also attested in Berom and N-V nominal compounds have an agentive or locative interpretation that is not solely determined by the meaning of the constituents of the compound but by the context of the usage.

Keywords: berom, berom compounds, nominal compound, N-V compounds

Procedia PDF Downloads 64