Search results for: neural machine translation

Authors: Guezouli Larbi, Belkacem Soundes

Abstract:

The automation of the text detection process can help the human in his driving task. Its application can be very useful to help drivers to have more information about their environment by facilitating the reading of road signs such as directional signs, events, stores, etc. In this paper, a system consisting of two stages has been proposed. In the first one, we used pseudo-Zernike moments to pinpoint areas of the image that may contain text. The architecture of this part is based on three main steps, region of interest (ROI) detection, text localization, and non-text region filtering. Then, in the second step, we present a convolutional neural network architecture (On-Road Text Detection Network - ORTDN) which is considered a classification phase. The results show that the proposed framework achieved ≈ 35 fps and an mAP of ≈ 90%, thus a low computational time with competitive accuracy.

Keywords: text detection, CNN, PZM, deep learning

Procedia PDF Downloads 70

Authors: Debora Gottardello, Mireia Valverde Aparicio, Juan Llopis Taverner

Abstract:

Globalization has given rise to a great diversity in the composition of people in organizations. Diversity management is therefore key to create growth in today’s competitive global marketplace. This work develops a literature review related to the existing models for managing diversity covering the period from 1980 until 2014. Furthermore, it identifies limitations in previous models. More specifically, the literature review reveals that there is a lack of information about how these models can be adapted from the headquarters to the subsidiaries. Therefore, the contribution of this paper is to suggest how the models should be adapted when they are directed to host countries. Our aim is to highlight the limitations of the developed models with regards to the translation of the diversity management practices to the subsidiaries. Accordingly, a model that will enable MNCs to ensure a global strategy is suggested. Taking advantage of the potential incorporated in a culturally diverse work team should be at the top of every international company’s aims. Executives from headquarters need to use different attitudes when transferring diversity practices towards their subsidiaries. Further studies should reassess local practices of diversity management to find out how this universal management model is translated.

Keywords: culture diversity, diversity management, human resources management, MNCs, subsidiaries, workforce diversity

Procedia PDF Downloads 239

Authors: Boris M. Fedorov, Liudmila P. Goncharenko, Sergey A. Sybachin, Natalia A. Mamedova, Ekaterina V. Makarenkova, Saule Rakhimova

Abstract:

In modern economic conditions, the question of the possibility of correct recognition of a signature on digital documents in order to verify the expression of will or confirm a certain operation is relevant. The additional complexity of processing lies in the dynamic variability of the signature for each individual, as well as in the way information is processed because the signature refers to biometric data. The article discusses the issues of using artificial intelligence models in order to improve the quality of signature confirmation in document recognition. The analysis of several possible options for using the model is carried out. The results of the study are given, in which it is possible to correctly determine the authenticity of the signature on small samples.

Keywords: signature recognition, biometric data, artificial intelligence, neural networks

Procedia PDF Downloads 132

Authors: Hadjoui Abdelhamid, Saimi Ahmed

Abstract:

The hybrid h-p finite element method for the dynamic behavior of nonlinear rotors is described in this paper. The standard h-version method of discretizing the problem is retained, but modified to allow the use of polynomially-enriched beam elements. A hierarchically enriching element will thus not affect the nodal displacement and rotation, but will influence the values of the nodal bending moment and shear force is used. The deterministic movements of rotation and translation of the support which are coupled to the excitations due to unbalance are also taken into account. We study also the geometric dissymmetry of the shaft and the disc, thus the equations of motion of the rotor contain variable parametric coefficients over time that can lead to a lateral dynamic instability. The effects of movements combined support for bearings are analyzed and discussed through Campbell diagrams and spectral analyses. A program is made in Matlab. After validation of the program, several examples are studied. The influence of physical and geometric parameters on the natural frequencies of the shaft is determined through the study of these examples. Among these parameters, we include the variation in the diameter and the thickness of the rotor, the position of the disc.

Keywords: Campbell diagram, critical speeds, nonlinear rotor, version h-p of FEM

Procedia PDF Downloads 221

Authors: Ling Liyun

Abstract:

In the field of contemporary architecture, complex forms of architectural works continue to emerge in the world, along with some new terminology emerged: digital architecture, parametric design, algorithm generation, building information modeling, CNC construction and so on. Architects gradually mastered the new skills of mathematical logic in the form of exploration, virtual simulation, and the entire design and coordination in the construction process. Digital construction technology has a greater degree in controlling construction, and ensure its accuracy, creating a series of new construction techniques. As a result, the use of digital technology is an improvement and expansion of the practice of digital architecture design revolution. We worked by reading and analyzing information about the digital architecture development process, a large number of cases, as well as architectural design and construction as a whole process. Thus current developments were introduced and discussed in our paper, such as architectural discourse, design theory, digital design models and techniques, material selecting, as well as artificial intelligence space design. Our paper also pays attention to the representative three cases of digital design and construction experiment at great length in detail to expound high-informatization, high-reliability intelligence, and high-technique in constructing a humane space to cope with the rapid development of urbanization. We concluded that the opportunities and challenges of the shift existed in architectural paradigms, such as the cooperation methods, theories, models, technologies and techniques which were currently employed in digital design research and digital praxis. We also find out that the innovative use of space can gradually change the way people learn, talk, and control information. The past two decades, digital technology radically breaks the technology constraints of industrial technical products, digests the publicity on a particular architectural style (era doctrine). People should not adapt to the machine, but in turn, it’s better to make the machine work for users.

Keywords: artificial intelligence, collaboration, digital architecture, digital design theory, material selection, space construction

Procedia PDF Downloads 123

Authors: Ebipatei Victoria Tunyan, T. A. Cao, Cheol Young Ock

Abstract:

Detecting subjectively biased statements is a vital task. This is because this kind of bias, when present in the text or other forms of information dissemination media such as news, social media, scientific texts, and encyclopedias, can weaken trust in the information and stir conflicts amongst consumers. Subjective bias detection is also critical for many Natural Language Processing (NLP) tasks like sentiment analysis, opinion identification, and bias neutralization. Having a system that can adequately detect subjectivity in text will boost research in the above-mentioned areas significantly. It can also come in handy for platforms like Wikipedia, where the use of neutral language is of importance. The goal of this work is to identify the subjectively biased language in text on a sentence level. With machine learning, we can solve complex AI problems, making it a good fit for the problem of subjective bias detection. A key step in this approach is to train a classifier based on BERT (Bidirectional Encoder Representations from Transformers) as upstream model. BERT by itself can be used as a classifier; however, in this study, we use BERT as data preprocessor as well as an embedding generator for a Bi-LSTM (Bidirectional Long Short-Term Memory) network incorporated with attention mechanism. This approach produces a deeper and better classifier. We evaluate the effectiveness of our model using the Wiki Neutrality Corpus (WNC), which was compiled from Wikipedia edits that removed various biased instances from sentences as a benchmark dataset, with which we also compare our model to existing approaches. Experimental analysis indicates an improved performance, as our model achieved state-of-the-art accuracy in detecting subjective bias. This study focuses on the English language, but the model can be fine-tuned to accommodate other languages.

Keywords: subjective bias detection, machine learning, BERT–BiLSTM–Attention, text classification, natural language processing

Procedia PDF Downloads 115

Authors: Kai Warsoenke, Maik Mackiewicz

Abstract:

To achieve a high quality of assembled car body structures, tolerancing is used to ensure a geometric accuracy of the single car body parts. There are two main techniques to determine the required tolerances. The first is tolerance analysis which describes the influence of individually tolerated input values on a required target value. Second is tolerance synthesis to determine the location of individual tolerances to achieve a target value. Both techniques are based on classical statistical methods, which assume certain probability distributions. To ensure competitiveness in both saturated and dynamic markets, production processes in vehicle manufacturing must be flexible and efficient. The dimensional specifications selected for the individual body components and the resulting assemblies have a major influence of the quality of the process. For example, in the manufacturing of forming tools as operating equipment or in the higher level of car body assembly. As part of the metrological process monitoring, manufactured individual parts and assemblies are recorded and the measurement results are stored in databases. They serve as information for the temporary adjustment of the production processes and are interpreted by experts in order to derive suitable adjustments measures. In the production of forming tools, this means that time-consuming and costly changes of the tool surface have to be made, while in the body shop, uncertainties that are difficult to control result in cost-intensive rework. The stored measurement results are not used to intelligently design tolerances in future processes or to support temporary decisions based on real-world geometric data. They offer potential to extend the tolerancing methods through data analysis and machine learning models. The purpose of this paper is to examine real-world measurement data from individual car body components, as well as assemblies, in order to develop an approach for using the data in short-term actions and future projects. For this reason, the measurement data will be analyzed descriptively in the first step in order to characterize their behavior and to determine possible correlations. In the following, a database is created that is suitable for developing machine learning models. The objective is to create an intelligent way to determine the position and number of measurement points as well as the local tolerance range. For this a number of different model types are compared and evaluated. The models with the best result are used to optimize equally distributed measuring points on unknown car body part geometries and to assign tolerance ranges to them. The current results of this investigation are still in progress. However, there are areas of the car body parts which behave more sensitively compared to the overall part and indicate that intelligent tolerancing is useful here in order to design and control preceding and succeeding processes more efficiently.

Keywords: automotive production, machine learning, process optimization, smart tolerancing

Procedia PDF Downloads 98

Authors: Moustafa Ibrahim Moustafa

Abstract:

The purpose of this study was to determine the immediate and long term effects of a multimodal program, with the addition of cervical sagittal curve restoration and forward head correction, on severity of dizziness, disability, frequency of dizziness, and severity of cervical pain. 72 patients with cervicogenic dizziness, definite hypolordotic cervical spine, and forward head posture were randomized to experimental or a control group. Both groups received the multimodal program, additionally, the study group received the Denneroll cervical traction. All outcome measures were measured at three intervals. The general linear model indicated a significant group × time effects in favor of experimental group on measures of anterior head translation (F=329.4 P < .0005), cervical lordosis (F=293.7 P < .0005), severity of dizziness (F=262.1 P < .0005), disability (F=248.9 P < .0005), frequency of dizziness (F=53.9 P < .0005), and severity of cervical pain (F=350.1 P < .0005). The addition of Dennroll cervical traction to a multimodal program can positively affect dizziness management outcomes.

Keywords: randomized controlled trial, traction, dizziness, cervical

Procedia PDF Downloads 289

Authors: Ole-Christian Galbo Engstrøm, Erik Schou Dreier, Birthe Møller Jespersen, Kim Steenstrup Pedersen

Abstract:

Grain quality analysis is a multi-parameterized problem that includes a variety of qualitative and quantitative parameters such as grain type classification, damage type classification, and nutrient regression. Currently, these parameters require human inspection, a multitude of instruments employing a variety of sensor technologies, and predictive model types or destructive and slow chemical analysis. This paper investigates the feasibility of applying near-infrared hyperspectral imaging (NIR-HSI) to grain quality analysis. For this study two datasets of NIR hyperspectral images in the wavelength range of 900 nm - 1700 nm have been used. Both datasets contain images of sparsely and densely packed grain kernels. The first dataset contains ~87,000 image crops of bulk wheat samples from 63 harvests where protein value has been determined by the FOSS Infratec NOVA which is the golden industry standard for protein content estimation in bulk samples of cereal grain. The second dataset consists of ~28,000 image crops of bulk grain kernels from seven different wheat varieties and a single rye variety. In the first dataset, protein regression analysis is the problem to solve while variety classification analysis is the problem to solve in the second dataset. Deep convolutional neural networks (CNNs) have the potential to utilize spatio-spectral correlations within a hyperspectral image to simultaneously estimate the qualitative and quantitative parameters. CNNs can autonomously derive meaningful representations of the input data reducing the need for advanced preprocessing techniques required for classical chemometric model types such as artificial neural networks (ANNs) and partial least-squares regression (PLS-R). A comparison between different CNN architectures utilizing 2D and 3D convolution is conducted. These results are compared to the performance of ANNs and PLS-R. Additionally, a variety of preprocessing techniques from image analysis and chemometrics are tested. These include centering, scaling, standard normal variate (SNV), Savitzky-Golay (SG) filtering, and detrending. The results indicate that the combination of NIR-HSI and CNNs has the potential to be the foundation for an automatic system unifying qualitative and quantitative grain quality analysis within a single sensor technology and predictive model type.

Keywords: deep learning, grain analysis, hyperspectral imaging, preprocessing techniques

Procedia PDF Downloads 82

Authors: Le Zhao, Alain Nogaret

Abstract:

We have built universal Central Pattern Generator (CPG) hardware by interconnecting Hodgkin-Huxley neurons with reciprocally inhibitory synapses. We investigate the dynamics of neuron oscillations as a function of the time delay between current steps applied to individual neurons. We demonstrate stimulus dependent switching between spiking polyrhythms and map the phase portraits of the neuron oscillations to reveal the basins of attraction of the system. We experimentally study the dependence of the attraction basins on the network parameters: the neuron response time and the strength of inhibitory connections.

Keywords: central pattern generator, winnerless competition principle, artificial neural networks, synapses

Procedia PDF Downloads 454

Authors: Samira Baghbanbari, A. B. P. Lever, Payam S. Shabestari, Donald Weaver

Abstract:

Existing signal transmission models, although undoubtedly useful, have proven insufficient to explain the full complexity of information transfer within the central nervous system. The development of transformative models will necessitate a more comprehensive understanding of neuronal lipid membrane electrophysiology. Pursuant to this goal, the role of highly organized interfacial phospholipid-water layers emerges as a promising case study. A series of phospholipids in neural-glial gap junction interfaces as well as cholesterol molecules have been computationally modelled using high-performance density functional theory (DFT) calculations. Subsequent 'charge decomposition analysis' calculations have revealed a net transfer of charge from phospholipid orbitals through the organized interfacial water layer before ultimately finding its way to cholesterol acceptor molecules. The specific pathway of charge transfer from phospholipid via water layers towards cholesterol has been mapped in detail. Cholesterol is an essential membrane component that is overrepresented in neuronal membranes as compared to other mammalian cells; given this relative abundance, its apparent role as an electronic acceptor may prove to be a relevant factor in further signal transmission studies of the central nervous system. The timescales over which this electronic charge transfer occurs have also been evaluated by utilizing a system design that systematically increases the number of water molecules separating lipids and cholesterol. Memory loss through hydrogen-bonded networks in water can occur at femtosecond timescales, whereas existing action potential-based models are limited to micro or nanosecond scales. As such, the development of future models that attempt to explain faster timescale signal transmission in the central nervous system may benefit from our work, which provides additional information regarding fast timescale energy transfer mechanisms occurring through interfacial water. The study possesses a dataset that includes six distinct phospholipids and a collection of cholesterol. Ten optimized geometric characteristics (features) were employed to conduct binary classification through an artificial neural network (ANN), differentiating cholesterol from the various phospholipids. This stems from our understanding that all lipids within the first group function as electronic charge donors, while cholesterol serves as an electronic charge acceptor.

Keywords: charge transfer, signal transmission, phospholipids, water layers, ANN

Procedia PDF Downloads 49

Authors: Aishrith Rao

Abstract:

COVID 19 is a pandemic which has resulted in thousands of deaths around the world and a huge impact on the global economy. Testing is a huge issue as the test kits have limited availability and are expensive to manufacture. Using deep learning methods on radiology images in the detection of the coronavirus as these images contain information about the spread of the virus in the lungs is extremely economical and time-saving as it can be used in areas with a lack of testing facilities. This paper focuses on binary classification and multi-class classification of COVID 19 and other diseases such as pneumonia, tuberculosis, etc. Different deep learning methods such as VGG-19, COVID-Net, ResNET+ SVM, Deep CNN, DarkCovidnet, etc., have been used, and their accuracy has been compared using the Chest X-Ray dataset.

Keywords: deep learning, computer vision, radiology, COVID-19, ResNet, VGG-19, deep neural networks

Procedia PDF Downloads 140

Authors: Amirreza Fazely Hamedani, Muzzamil Aziz, Philipp Wieder, Ramin Yahyapour

Abstract:

Software-defined networking in recent years came into the sight of so many network designers as a successor to the traditional networking. Unlike traditional networks where control and data planes engage together within a single device in the network infrastructure such as switches and routers, the two planes are kept separated in software-defined networks (SDNs). All critical decisions about packet routing are made on the network controller, and the data level devices forward the packets based on these decisions. This type of network is vulnerable to DDoS attacks, degrading the overall functioning and performance of the network by continuously injecting the fake flows into it. This increases substantial burden on the controller side, and the result ultimately leads to the inaccessibility of the controller and the lack of network service to the legitimate users. Thus, the protection of this novel network architecture against denial of service attacks is essential. In the world of cybersecurity, attacks and new threats emerge every day. It is essential to have tools capable of managing and analyzing all this new information to detect possible attacks in real-time. These tools should provide a comprehensive solution to automatically detect, predict and prevent abnormalities in the network. Big data encompasses a wide range of studies, but it mainly refers to the massive amounts of structured and unstructured data that organizations deal with on a regular basis. On the other hand, it regards not only the volume of the data; but also that how data-driven information can be used to enhance decision-making processes, security, and the overall efficiency of a business. This paper presents an intelligent big data framework as a solution to handle illegitimate traffic burden on the SDN network created by the numerous DDoS attacks. The framework entails an efficient defence and monitoring mechanism against DDoS attacks by employing the state of the art machine learning techniques.

Keywords: apache spark, apache kafka, big data, DDoS attack, machine learning, SDN network

Procedia PDF Downloads 156

Authors: Marc Toutain, Pascale Leconte, Antoine Gauthier

Abstract:

Introduction: Eating Disorders (ED), such as anorexia, bulimia, and binge eating, are psychiatric illnesses that mostly affect young people. The main symptoms concern eating (restriction, excessive food intake) and weight control behaviors (laxatives, vomiting). Psychological comorbidities (depression, executive function disorders, etc.) and problematic behaviors toward physical activity (PA) are commonly associated with ED. Acquaintances on ED risk factors are still lacking, and more community sample studies are needed to improve prevention and early detection. To our knowledge, studies are needed to specifically investigate the link between ED risk level, PA, and psychological risk factors in a community sample of adolescents. The aim of this study is to assess the relation between ED risk level, exercise (type, frequency, and motivations for engaging in exercise), and psychological factors based on the Jacobi risk factors model. We suppose that a high risk of ED will be associated with the practice of high caloric cost PA, motivations oriented to weight and shape control, and psychological disturbances. Method: An online survey destined for students has been sent to several middle schools and colleges in northwest France. This survey combined several questionnaires, the Eating Attitude Test-26 assessing ED risk; the Exercise Motivation Inventory–2 assessing motivations toward PA; the Hospital Anxiety and Depression Scale assessing anxiety and depression, the Contour Drawing Rating Scale; and the Body Esteem Scale assessing body dissatisfaction, Rosenberg Self-esteem Scale assessing self-esteem, the Exercise Dependence Scale-Revised assessing PA dependence, the Multidimensional Assessment of Interoceptive Awareness assessing interoceptive awareness and the Frost Multidimensional Perfectionism Scale assessing perfectionism. Machine learning analysis will be performed in order to constitute groups with a tree-based model clustering method, extract risk profile(s) with a bootstrap method comparison, and predict ED risk with a prediction method based on a decision tree-based model. Expected results: 1044 complete records have already been collected, and the survey will be closed at the end of May 2022. Records will be analyzed with a clustering method and a bootstrap method in order to reveal risk profile(s). Furthermore, a predictive tree decision method will be done to extract an accurate predictive model of ED risk. This analysis will confirm typical main risk factors and will give more data on presumed strong risk factors such as exercise motivations and interoceptive deficit. Furthermore, it will enlighten particular risk profiles with a strong level of proof and greatly contribute to improving the early detection of ED and contribute to a better understanding of ED risk factors.

Keywords: eating disorders, risk factors, physical activity, machine learning

Procedia PDF Downloads 70

Authors: Lina, Arlends Chris, Bagus Mulyawan, Agus B. Dharmawan

Abstract:

Blood cell analysis plays a significant role in the diagnosis of human health. As an alternative to the traditional technique conducted by laboratory technicians, this paper presents an automatic white blood cell (leukocyte) detection system using Image Stitching and Color Overlapping Windows. The advantage of this method is to present a detection technique of white blood cells that are robust to imperfect shapes of blood cells with various image qualities. The input for this application is images from a microscope-slide translation video. The preprocessing stage is performed by stitching the input images. First, the overlapping parts of the images are determined, then stitching and blending processes of two input images are performed. Next, the Color Overlapping Windows is performed for white blood cell detection which consists of color filtering, window candidate checking, window marking, finds window overlaps, and window cropping processes. Experimental results show that this method could achieve an average of 82.12% detection accuracy of the leukocyte images.

Keywords: color overlapping windows, image stitching, leukocyte detection, white blood cell detection

Procedia PDF Downloads 294

Authors: O. J. Akinsola, Yinan Zheng, Rose Anorlu, F. T. Ogunsola, Lifang Hou, Robert Leo-Murphy

Abstract:

Cervical Cancer (CC) is the 2nd most common cancer among women living in low and middle-income countries, with no associated symptoms during formative periods. With the advancement and innovative medical research, there are numerous preventive measures being utilized, but the incidence of cervical cancer cannot be truncated with the application of only screening tests. The mortality associated with this invasive cervical cancer can be nipped in the bud through the important role of early-stage detection. This study research selected an array of different top features selection techniques which was aimed at developing a model that could validly diagnose the risk factors of cervical cancer. A retrospective clinic-based cohort study was conducted on 178 HIV-associated cervical cancer patients in Lagos University teaching Hospital, Nigeria (U54 data repository) in April 2022. The outcome measure was the automated prediction of the HIV-associated cervical cancer cases, while the predictor variables include: demographic information, reproductive history, birth control, sexual history, cervical cancer screening history for invasive cervical cancer. The proposed technique was assessed with R and Python programming software to produce the model by utilizing the classification algorithms for the detection and diagnosis of cervical cancer disease. Four machine learning classification algorithms used are: the machine learning model was split into training and testing dataset into ratio 80:20. The numerical features were also standardized while hyperparameter tuning was carried out on the machine learning to train and test the data. Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), and K-Nearest Neighbor (KNN). Some fitting features were selected for the detection and diagnosis of cervical cancer diseases from selected characteristics in the dataset using the contribution of various selection methods for the classification cervical cancer into healthy or diseased status. The mean age of patients was 49.7±12.1 years, mean age at pregnancy was 23.3±5.5 years, mean age at first sexual experience was 19.4±3.2 years, while the mean BMI was 27.1±5.6 kg/m2. A larger percentage of the patients are Married (62.9%), while most of them have at least two sexual partners (72.5%). Age of patients (OR=1.065, p<0.001**), marital status (OR=0.375, p=0.011**), number of pregnancy live-births (OR=1.317, p=0.007**), and use of birth control pills (OR=0.291, p=0.015**) were found to be significantly associated with HIV-associated cervical cancer. On top ten 10 features (variables) considered in the analysis, RF claims the overall model performance, which include: accuracy of (72.0%), the precision of (84.6%), a recall of (84.6%) and F1-score of (74.0%) while LR has: an accuracy of (74.0%), precision of (70.0%), recall of (70.0%) and F1-score of (70.0%). The RF model identified 10 features predictive of developing cervical cancer. The age of patients was considered as the most important risk factor, followed by the number of pregnancy livebirths, marital status, and use of birth control pills, The study shows that data mining techniques could be used to identify women living with HIV at high risk of developing cervical cancer in Nigeria and other sub-Saharan African countries.

Keywords: associated cervical cancer, data mining, random forest, logistic regression

Procedia PDF Downloads 69

Authors: Nir Nissim, Erez Shalom, Tomer Lancewiki, Yuval Elovici, Yuval Shahar

Abstract:

Background: A Medical Device (MD) is an instrument, machine, implant, or similar device that includes a component intended for the purpose of the diagnosis, cure, treatment, or prevention of disease in humans or animals. Medical devices play increasingly important roles in health services eco-systems, including: (1) Patient Diagnostics and Monitoring; Medical Treatment and Surgery; and Patient Life Support Devices and Stabilizers. MDs are part of the medical device eco-system and are connected to the network, sending vital information to the internal medical information systems of medical centers that manage this data. Wireless components (e.g. Wi-Fi) are often embedded within medical devices, enabling doctors and technicians to control and configure them remotely. All these functionalities, roles, and uses of MDs make them attractive targets of cyber-attacks launched for many malicious goals; this trend is likely to significantly increase over the next several years, with increased awareness regarding MD vulnerabilities, the enhancement of potential attackers’ skills, and expanded use of medical devices. Significance: We propose to develop and implement Cyber-Med, a unique collaborative project of Ben-Gurion University of the Negev and the Clalit Health Services Health Maintenance Organization. Cyber-Med focuses on the development of a comprehensive detection framework that relies on a critical attack repository that we aim to create. Cyber-Med will allow researchers and companies to better understand the vulnerabilities and attacks associated with medical devices as well as providing a comprehensive platform for developing detection solutions. Methodology: The Cyber-Med detection framework will consist of two independent, but complementary detection approaches: one for known attacks, and the other for unknown attacks. These modules incorporate novel ideas and algorithms inspired by our team's domains of expertise, including cyber security, biomedical informatics, and advanced machine learning, and temporal data mining techniques. The establishment and maintenance of Cyber-Med’s up-to-date attack repository will strengthen the capabilities of Cyber-Med’s detection framework. Major Findings: Based on our initial survey, we have already found more than 15 types of vulnerabilities and possible attacks aimed at MDs and their eco-system. Many of these attacks target individual patients who use devices such pacemakers and insulin pumps. In addition, such attacks are also aimed at MDs that are widely used by medical centers such as MRIs, CTs, and dialysis engines; the information systems that store patient information; protocols such as DICOM; standards such as HL7; and medical information systems such as PACS. However, current detection tools, techniques, and solutions generally fail to detect both the known and unknown attacks launched against MDs. Very little research has been conducted in order to protect these devices from cyber-attacks, since most of the development and engineering efforts are aimed at the devices’ core medical functionality, the contribution to patients’ healthcare, and the business aspects associated with the medical device.

Keywords: medical device, cyber security, attack, detection, machine learning

Procedia PDF Downloads 345

Authors: Kishor Chandra Kandpal, Amit Kumar

Abstract:

The Indian Himalaya region harbours approximately 1748 plants of medicinal importance, and as per International Union for Conservation of Nature (IUCN), the 112 plant species among these are threatened and endangered. To ease the pressure on these plants, the government of India is encouraging its in-situ cultivation. The Saussurea costus, Valeriana jatamansi, and Picrorhiza kurroa have also been prioritized for large scale cultivation owing to their market demand, conservation value and medicinal properties. These species are found from 1000 m to 4000 m elevation ranges in the Indian Himalaya. Identification of these plants in the field requires taxonomic skills, which is one of the major bottleneck in the conservation and management of these plants. In recent years, Hyperspectral remote sensing techniques have been precisely used for the discrimination of plant species with the help of their unique spectral signatures. In this background, a spectral library of the above 03 medicinal plants was prepared by collecting the spectral data using a handheld spectroradiometer (325 to 1075 nm) from farmer’s fields of Himachal Pradesh and Uttarakhand states of Indian Himalaya. The Random forest (RF) model was implied on the spectral data for the classification of the medicinal plants. The 80:20 standard split ratio was followed for training and validation of the RF model, which resulted in training accuracy of 84.39 % (kappa coefficient = 0.72) and testing accuracy of 85.29 % (kappa coefficient = 0.77). This RF classifier has identified green (555 to 598 nm), red (605 nm), and near-infrared (725 to 840 nm) wavelength regions suitable for the discrimination of these species. The findings of this study have provided a technique for rapid and onsite identification of the above medicinal plants in the field. This will also be a key input for the classification of hyperspectral remote sensing images for mapping of these species in farmer’s field on a regional scale. This is a pioneer study in the Indian Himalaya region for medicinal plants in which the applicability of hyperspectral remote sensing has been explored.

Keywords: himalaya, hyperspectral remote sensing, machine learning; medicinal plants, random forests

Procedia PDF Downloads 191

Authors: Rahul Paul, Peter Mctaggart, Luke Skinner

Abstract:

Maintaining public safety and network reliability are the core objectives of all electricity distributors globally. For many electricity distributors, managing vegetation clearances from their above ground assets (poles and conductors) is the most important and costly risk mitigation control employed to meet these objectives. Light Detection And Ranging (LiDAR) is widely used by utilities as a cost-effective method to inspect their spatially-distributed assets at scale, often captured using high powered LiDAR scanners attached to fixed wing or rotary aircraft. The resulting 3D point cloud model is used by these utilities to perform engineering grade measurements that guide the prioritisation of vegetation cutting programs. Advances in computer vision and machine-learning approaches are increasingly applied to increase automation and reduce inspection costs and time; however, real-world LiDAR capture variables (e.g., aircraft speed and height) create complexity, noise, and missing data, reducing the effectiveness of these approaches. This paper proposes a method for identifying each conductor from LiDAR data via clustering methods that can precisely reconstruct conductors in complex real-world configurations in the presence of high levels of noise. It proposes 3D catenary models for individual clusters fitted to the captured LiDAR data points using a least square method. An iterative learning process is used to identify potential conductor models between pole pairs. The proposed method identifies the optimum parameters of the catenary function and then fits the LiDAR points to reconstruct the conductors.

Keywords: point cloud, LİDAR data, machine learning, computer vision, catenary curve, vegetation management, utility industry

Procedia PDF Downloads 84

Authors: C. Jameke

Abstract:

During the cognitivist era in cognitive psychology, a theory of internal rules and symbolic representations was posited as an account of human cognition. This type of cognitive architecture had its heyday during the 1970s and 80s, but it has now been largely abandoned in favour of subsymbolic architectures (e.g. connectionism), non-representational frameworks (e.g. dynamical systems theory), and statistical approaches such as Bayesian theory. In this presentation I describe this changing landscape of research, and comment on the increasing influence of neuroscience on cognitive psychology. I then briefly review a few recent developments in connectionism, and neurocomputation relevant to cognitive psychology, and critically discuss the assumption made by some researchers in these frameworks that higher-level aspects of human cognition are simply emergent properties of massively large distributed neural networks

Keywords: connectionism, emergentism, postocgnitivist, representations, subsymbolic archiitecture

Procedia PDF Downloads 551

Authors: Fernando Ojedo, Alejandro Alvarez, Pedro Macizo

Abstract:

In the present study, we explored electrophysiological correlates of Arabic digits naming to determine semantic processing of numbers. Participants named Arabic digits grouped by category or intermixed with exemplars of other semantic categories while the N400 event-related potential was examined. Around 350-450 ms after the presentation of Arabic digits, brain waves were more positive in anterior regions and more negative in posterior regions when stimuli were grouped by category relative to the mixed condition. Contrary to what was found in other studies, electrophysiological results suggested that the production of numerals involved semantic mediation.

Keywords: Arabic digit naming, event-related potentials, semantic processing, number production

Procedia PDF Downloads 566

Authors: Ammar Alali, Mahmoud Abughaban, William Contreras Otalvora

Abstract:

Optimizing the drilling process for cost and efficiency requires the optimization of the rate of penetration (ROP). ROP is the measurement of the speed at which the wellbore is created, in units of feet per hour. It is the primary indicator of measuring drilling efficiency. Maximization of the ROP can indicate fast and cost-efficient drilling operations; however, high ROPs may induce unintended events, which may lead to nonproductive time (NPT) and higher net costs. The proposed ROP optimization solution is a hybrid, data-driven system that aims to improve the drilling process, maximize the ROP, and minimize NPT. The system consists of two phases: (1) utilizing existing geological and drilling data to train the model prior, and (2) real-time adjustments of the controllable dynamic drilling parameters [weight on bit (WOB), rotary speed (RPM), and pump flow rate (GPM)] that direct influence on the ROP. During the first phase of the system, geological and historical drilling data are aggregated. After, the top-rated wells, as a function of high instance ROP, are distinguished. Those wells are filtered based on NPT incidents, and a cross-plot is generated for the controllable dynamic drilling parameters per ROP value. Subsequently, the parameter values (WOB, GPM, RPM) are calculated as a conditioned mean based on physical distance, following Inverse Distance Weighting (IDW) interpolation methodology. The first phase is concluded by producing a model of drilling best practices from the offset wells, prioritizing the optimum ROP value. This phase is performed before the commencing of drilling. Starting with the model produced in phase one, the second phase runs an automated drill-off test, delivering live adjustments in real-time. Those adjustments are made by directing the driller to deviate two of the controllable parameters (WOB and RPM) by a small percentage (0-5%), following the Constrained Random Search (CRS) methodology. These minor incremental variations will reveal new drilling conditions, not explored before through offset wells. The data is then consolidated into a heat-map, as a function of ROP. A more optimum ROP performance is identified through the heat-map and amended in the model. The validation process involved the selection of a planned well in an onshore oil field with hundreds of offset wells. The first phase model was built by utilizing the data points from the top-performing historical wells (20 wells). The model allows drillers to enhance decision-making by leveraging existing data and blending it with live data in real-time. An empirical relationship between controllable dynamic parameters and ROP was derived using Artificial Neural Networks (ANN). The adjustments resulted in improved ROP efficiency by over 20%, translating to at least 10% saving in drilling costs. The novelty of the proposed system lays is its ability to integrate historical data, calibrate based geological formations, and run real-time global optimization through CRS. Those factors position the system to work for any newly drilled well in a developing field event.

Keywords: drilling optimization, geological formations, machine learning, rate of penetration

Procedia PDF Downloads 113

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 154

Authors: Kumari Sweta, Ajanta Goswami, Abhilasha Dixit

Abstract:

Landslide is a geomorphic process that plays an essential role in the evolution of the hill-slope and long-term landscape evolution. But its abrupt nature and the associated catastrophic forces of the process can have undesirable socio-economic impacts, like substantial economic losses, fatalities, ecosystem, geomorphologic and infrastructure disturbances. The estimated fatality rate is approximately 1person /100 sq. Km and the average economic loss is more than 550 crores/year in the Himalayan belt due to landslides. This study presents a comparative performance of a statistical bivariate method and a machine learning technique for landslide susceptibility mapping in and around Dharamshala, Himachal Pradesh. The final produced landslide susceptibility maps (LSMs) with better accuracy could be used for land-use planning to prevent future losses. Dharamshala, a part of North-western Himalaya, is one of the fastest-growing tourism hubs with a total population of 30,764 according to the 2011 census and is amongst one of the hundred Indian cities to be developed as a smart city under PM’s Smart Cities Mission. A total of 209 landslide locations were identified in using high-resolution linear imaging self-scanning (LISS IV) data. The thematic maps of parameters influencing landslide occurrence were generated using remote sensing and other ancillary data in the GIS environment. The landslide causative parameters used in the study are slope angle, slope aspect, elevation, curvature, topographic wetness index, relative relief, distance from lineaments, land use land cover, and geology. LSMs were prepared using information value (Info Val), and Fuzzy Expert System (FES) models. Info Val is a statistical bivariate method, in which information values were calculated as the ratio of the landslide pixels per factor class (Si/Ni) to the total landslide pixel per parameter (S/N). Using this information values all parameters were reclassified and then summed in GIS to obtain the landslide susceptibility index (LSI) map. The FES method is a machine learning technique based on ‘mean and neighbour’ strategy for the construction of fuzzifier (input) and defuzzifier (output) membership function (MF) structure, and the FR method is used for formulating if-then rules. Two types of membership structures were utilized for membership function Bell-Gaussian (BG) and Trapezoidal-Triangular (TT). LSI for BG and TT were obtained applying membership function and if-then rules in MATLAB. The final LSMs were spatially and statistically validated. The validation results showed that in terms of accuracy, Info Val (83.4%) is better than BG (83.0%) and TT (82.6%), whereas, in terms of spatial distribution, BG is best. Hence, considering both statistical and spatial accuracy, BG is the most accurate one.

Keywords: bivariate statistical techniques, BG and TT membership structure, fuzzy expert system, information value method, machine learning technique

Procedia PDF Downloads 113

Authors: Waqas Ashraf, Yasir Nawab, Haritham Khan, Habib Awais, Shahbaz Ahmad

Abstract:

Seersucker is a popular lightweight fabric widely used in men’s and women’s suiting, casual wear, children’s clothing, house robes, bed spreads and for spring and summer wear. The puckered effect generates air spaces between body and the fabric, keeping the wearer cool in hot conditions. The aim of this work was to develop knitted seersucker fabric on single cylinder weft knitting machine using plain jersey structure. Core spun cotton yarn and cotton spun yarn of same linear density were used. Core spun cotton yarn, contains cotton fiber in the sheath and elastase filament in the core. The both yarn were fed at regular interval to feeders on the machine. The loop length and yarn tension were kept constant at each feeder. The samples were then scoured and bleached. After wet processing, the fabric samples were washed and tumble dried. Parameters like loop length, stitch density and areal density were measured after conditioning these samples for 24 hours in Standard atmospheric condition. Produced sample has a regular puckering stripe along the width of the fabric with same height. The stitch density of both the flat and puckered area of relaxed fabric was found to be different .Air permeability and moisture management tests were performed. The results indicated that the knitted seersucker fabric has better wicking and moisture management properties as the flat area contact, whereas puckered area held away from the skin. Seersucker effect in knitted fabric was achieved by the difference of contraction of both sets of courses produced from different types of yarns. The seer sucker fabric produce by knitting technique is less expensive as compared to woven seer sucker fabric as there is no need of yarn preparation. The knitted seersucker fabric is more practicable for summer dresses, skirts, blouses, shirts, trousers and shorts.

Keywords: air permeability, knitted structure, moisture management, seersucker

Procedia PDF Downloads 309

Authors: Joseph Bertrand Iboum Kissaaka, Charles Fonyuy Ngum Tchioben, Paul Gustave Fowe Kwetche, Jeannette Ngo Elogan Ntem, Joseph Binyet Njebakal, Ribert Yvan Makosso-Tchapi, François Mvondo Owono, Marie Joseph Ntamak-Nida

Abstract:

Located in the Gulf of Guinea, the Kribi-Campo sub-basin belongs to the Aptian salt basins along the West African Margin. In this paper, we investigated the tectono-stratigraphic architecture of the basin, focusing on the role of salt tectonics and strike-slip faults along the Kribi Fracture Zone with implications for reservoir prediction. Using 2D seismic data and well data interpreted through sequence stratigraphy with integrated seismic attributes analysis with Python Programming and unsupervised Machine Learning, at least six second-order sequences, indicating three main stages of tectono-stratigraphic evolution, were determined: pre-salt syn-rift, post-salt rift climax and post-rift stages. The pre-salt syn-rift stage with KTS1 tectonosequence (Barremian-Aptian) reveals a transform rifting along NE-SW transfer faults associated with N-S to NNE-SSW syn-rift longitudinal faults bounding a NW-SE half-graben filled with alluvial to lacustrine-fan delta deposits. The post-salt rift-climax stage (Lower to Upper Cretaceous) includes two second-order tectonosequences (KTS2 and KTS3) associated with the salt tectonics and Campo High uplift. During the rift-climax stage, the growth of salt diapirs developed syncline withdrawal basins filled by early forced regression, mid transgressive and late normal regressive systems tracts. The early rift climax underlines some fine-grained hangingwall fans or delta deposits and coarse-grained fans from the footwall of fault scarps. The post-rift stage (Paleogene to Neogene) contains at least three main tectonosequences KTS4, KTS5 and KTS6-7. The first one developed some turbiditic lobe complexes considered as mass transport complexes and feeder channel-lobe complexes cutting the unstable shelf edge of the Campo High. The last two developed submarine Channel Complexes associated with lobes towards the southern part and braided delta to tidal channels towards the northern part of the Kribi-Campo sub-basin. The reservoir distribution in the Kribi-Campo sub-basin reveals some channels, fan lobes reservoirs and stacked channels reaching up to the polygonal fault systems.

Keywords: tectono-stratigraphic architecture, Kribi-Campo sub-basin, machine learning, pre-salt sequences, post-salt sequences

Procedia PDF Downloads 31

Authors: Ramon Santana

Abstract:

The use of biometric identifiers in the field of information security, access control to resources, authentication in ATMs and banking among others, are of great concern because of the safety of biometric data. In the general architecture of a biometric system have been detected eight vulnerabilities, six of them allow obtaining minutiae template in plain text. The main consequence of obtaining minutia templates is the loss of biometric identifier for life. To mitigate these vulnerabilities several models to protect minutiae templates have been proposed. Several vulnerabilities in the cryptographic security of these models allow to obtain biometric data in plain text. In order to increase the cryptographic security and ease of reversibility, a minutiae templates protection model is proposed. The model aims to make the cryptographic protection and facilitate the reversibility of data using two levels of security. The first level of security is the data transformation level. In this level generates invariant data to rotation and translation, further transformation is irreversible. The second level of security is the evaluation level, where the encryption key is generated and data is evaluated using a defined evaluation function. The model is aimed at mitigating known vulnerabilities of the proposed models, basing its security on the impossibility of the polynomial reconstruction.

Keywords: fingerprint, template protection, bio-cryptography, minutiae protection

Procedia PDF Downloads 159

Authors: Dona Abdulwassi, Dhaee Dahlawi, Yara Zainy, Leen Joharji

Abstract:

The relationship between Netflix and artificial intelligence is discussed in this paper. Netflix uses the most effective and efficient approaches to apply artificial intelligence, machine learning, and data science. Netflix employs the personalization tool for their users, recommending or suggesting shows based on what those users have already watched. The researchers conducted an experiment to learn more about how Netflix is used and how AI affects the user experience. The main conclusions of this study are that Netflix has a wide range of AI features, most users are happy with their Netflix subscriptions, and the majority prefer Netflix to alternative apps.

Keywords: easy accessibility, recommends, accuracy, privacy

Procedia PDF Downloads 51

Authors: Marcin Baranski

Abstract:

This article presents a vibration diagnostic method designed for permanent magnets (PM) traction motors. Those machines are commonly used in traction drives of electrical vehicles. Specific structural properties of machines excited by permanent magnets are used in this method - electromotive force (EMF) generated due to vibrations. This work presents: field-circuit model, results of static tests, results of calculations and simulations.

Keywords: electrical vehicle, permanent magnet, traction drive, vibrations, electrical machine, eccentricity

Procedia PDF Downloads 611

Authors: Okba Belahssen, Said Benramache

Abstract:

This paper presents wear behavior of the plasma-nitrided 20MnCr5 steel. Untreated and plasma nitrided samples were tested. The morphology was observed by scanning electron microscopy (SEM). The plasma nitriding behaviors of 20MnCr5 steel have been assessed by evaluating tribological properties and surface hardness by using a pin-on-disk wear machine and microhardness tester. Experimental results showed that the nitrides ε-Fe2−3N and γ′-Fe4N present in the white layer improve the wear resistance.

Keywords: plasma-nitriding, alloy 20mncr5, steel, friction, wear

Procedia PDF Downloads 537