Search results for: graph convolutional networks

Authors: İsmail Kavdır, M. Burak Büyükcan, Ferhat Kurtulmuş

Abstract:

Table olive is a valuable product especially in Mediterranean countries. It is usually consumed after some fermentation process. Defects happened naturally or as a result of an impact while olives are still fresh may become more distinct after processing period. Defected olives are not desired both in table olive and olive oil industries as it will affect the final product quality and reduce market prices considerably. Therefore it is critical to sort table olives before processing or even after processing according to their quality and surface defects. However, doing manual sorting has many drawbacks such as high expenses, subjectivity, tediousness and inconsistency. Quality criterions for green olives were accepted as color and free of mechanical defects, wrinkling, surface blemishes and rotting. In this study, it was aimed to classify fresh table olives using different classifiers and NIR spectroscopy readings and also to compare the classifiers. For this purpose, green (Ayvalik variety) olives were classified based on their surface feature properties such as defect-free, with bruised defect and with fly defect using FT-NIR spectroscopy and classification algorithms such as artificial neural networks, ident and cluster. Bruker multi-purpose analyzer (MPA) FT-NIR spectrometer (Bruker Optik, GmbH, Ettlingen Germany) was used for spectral measurements. The spectrometer was equipped with InGaAs detectors (TE-InGaAs internal for reflectance and RT-InGaAs external for transmittance) and a 20-watt high intensity tungsten–halogen NIR light source. Reflectance measurements were performed with a fiber optic probe (type IN 261) which covered the wavelengths between 780–2500 nm, while transmittance measurements were performed between 800 and 1725 nm. Thirty-two scans were acquired for each reflectance spectrum in about 15.32 s while 128 scans were obtained for transmittance in about 62 s. Resolution was 8 cm⁻¹ for both spectral measurement modes. Instrument control was done using OPUS software (Bruker Optik, GmbH, Ettlingen Germany). Classification applications were performed using three classifiers; Backpropagation Neural Networks, ident and cluster classification algorithms. For these classification applications, Neural Network tool box in Matlab, ident and cluster modules in OPUS software were used. Classifications were performed considering different scenarios; two quality conditions at once (good vs bruised, good vs fly defect) and three quality conditions at once (good, bruised and fly defect). Two spectrometer readings were used in classification applications; reflectance and transmittance. Classification results obtained using artificial neural networks algorithm in discriminating good olives from bruised olives, from olives with fly defect and from the olive group including both bruised and fly defected olives with success rates respectively changing between 97 and 99%, 61 and 94% and between 58.67 and 92%. On the other hand, classification results obtained for discriminating good olives from bruised ones and also for discriminating good olives from fly defected olives using the ident method ranged between 75-97.5% and 32.5-57.5%, respectfully; results obtained for the same classification applications using the cluster method ranged between 52.5-97.5% and between 22.5-57.5%.

Keywords: artificial neural networks, statistical classifiers, NIR spectroscopy, reflectance, transmittance

Procedia PDF Downloads 249

Authors: Maciej Jedrzejczyk, Karolina Marzantowicz

Abstract:

Traditional utility value chain is being transformed during last few years into unbundled markets. Increased distributed generation of energy is one of considerable challenges faced by Smart Grid networks. New sources of energy introduce volatile demand response which has a considerable impact on traditional middlemen in E&U market. The purpose of this research is to search for ways to allow near-real-time electricity markets to transact with surplus energy based on accurate time synchronous measurements. A proposed framework evaluates the use of secure peer-2-peer (P2P) communication and distributed transaction ledgers to provide flat hierarchy, and allow real-time insights into present and forecasted grid operations, as well as state and health of the network. An objective is to achieve dynamic grid operations with more efficient resource usage, higher security of supply and longer grid infrastructure life cycle. Methods used for this study are based on comparative analysis of different distributed ledger technologies in terms of scalability, transaction performance, pluggability with external data sources, data transparency, privacy, end-to-end security and adaptability to various market topologies. An intended output of this research is a design of a framework for safer, more efficient and scalable Smart Grid network which is bridging a gap between traditional components of the energy network and individual energy producers. Results of this study are ready for detailed measurement testing, a likely follow-up in separate studies. New platforms for Smart Grid achieving measurable efficiencies will allow for development of new types of Grid KPI, multi-smart grid branches, markets, and businesses.

Keywords: autonomous agents, Distributed computing, distributed ledger technologies, large scale systems, micro grids, peer-to-peer networks, Self-organization, self-stabilization, smart grids

Procedia PDF Downloads 306

Authors: Agastya Pratap Singh

Abstract:

Generative Adversarial Networks (GANs) have emerged as powerful tools in the fields of image recognition and anomaly detection due to their ability to model complex data distributions and generate realistic images. This systematic review explores recent advancements and applications of GANs in both image recognition and anomaly detection tasks. We discuss various GAN architectures, such as DCGAN, CycleGAN, and StyleGAN, which have been tailored to improve accuracy, robustness, and efficiency in visual data analysis. In image recognition, GANs have been used to enhance data augmentation, improve classification models, and generate high-quality synthetic images. In anomaly detection, GANs have proven effective in identifying rare and subtle abnormalities across various domains, including medical imaging, cybersecurity, and industrial inspection. The review also highlights the challenges and limitations associated with GAN-based methods, such as instability during training and mode collapse, and suggests future research directions to overcome these issues. Through this review, we aim to provide researchers with a comprehensive understanding of the capabilities and potential of GANs in transforming image recognition and anomaly detection practices.

Keywords: generative adversarial networks, image recognition, anomaly detection, DCGAN, CycleGAN, StyleGAN, data augmentation

Procedia PDF Downloads 28

Authors: Marzieh Ahmadi, Ruhullah Alikhan Gorgani

Abstract:

The usual methods of measuring regional inequalities can not reflect the internal changes of the country in terms of their displacement in different development groups, and the indicators of inequalities are not effective in demonstrating the dynamics of the distribution of inequality. For this purpose, this paper examines the dynamics of the urban inertial transport in the country during the period of 2006-2016 using the CIRD multidimensional index and stochastic kernel density method. it firstly selects 25 indicators in five dimensions including macroeconomic conditions, science and innovation, environmental sustainability, human capital and public facilities, and two-stage Principal Component Analysis methodology are developed to create a composite index of inequality. Then, in the second stage, using a nonparametric analytical approach to internal distribution dynamics and a stochastic kernel density method, the convergence hypothesis of the CIRD index of the Iranian provinces center is tested, and then, based on the ergodic density, long-run equilibrium is shown. Also, at this stage, for the purpose of adopting accurate regional policies, the distribution dynamics and process of convergence or divergence of the Iranian provinces for each of the five. According to the results of the first Stage, in 2006 & 2016, the highest level of development is related to Tehran and zahedan is at the lowest level of development. The results show that the central cities of the country are at the highest level of development due to the effects of Tehran's knowledge spillover and the country's lower cities are at the lowest level of development. The main reason for this may be the lack of access to markets in the border provinces. Based on the results of the second stage, which examines the dynamics of regional inequality transmission in the country during 2006-2016, the first year (2006) is not multifaceted and according to the kernel density graph, the CIRD index of about 70% of the cities. The value is between -1.1 and -0.1. The rest of the sequence on the right is distributed at a level higher than -0.1. In the kernel distribution, a convergence process is observed and the graph points to a single peak. Tends to be a small peak at about 3 but the main peak at about-0.6. According to the chart in the final year (2016), the multidimensional pattern remains and there is no mobility in the lower level groups, but at the higher level, the CIRD index accounts for about 45% of the provinces at about -0.4 Take it. That this year clearly faces the twin density pattern, which indicates that the cities tend to be closely related to each other in terms of development, so that the cities are low in terms of development. Also, according to the distribution dynamics results, the provinces of Iran follow the single-density density pattern in 2006 and the double-peak density pattern in 2016 at low and moderate inequality index levels and also in the development index. The country diverges during the years 2006 to 2016.

Keywords: Urban Disparity, CIRD Index, Convergence, Distribution Dynamics, Random Kernel Density

Procedia PDF Downloads 126

Authors: Nureni Asafe Yekini, Olawale Nasiru Lawal, Bola Dada, Gabriel Adeyemi Okunlola

Abstract:

This study is focused on the strengths and challenges associated with using the online social network as a rapidly evolving medium in marketing tourism services and businesses among the youths in Nigeria. The paper examines the Nigerian tourists’ attitude, mainly towards three aspects: application of Internet for travel and tourism; usage of online social networks in sharing travel and tourism experiences; and trust in electronic-media for marketing tourism businesses and services. The aim of this research is to determine the level of application of internet tools in marketing tourism businesses and services in Nigeria. This study reports an empirical analysis based on data obtained from a survey among 1004 Nigerian tourists. The outcome confirms the research hypothesis and points to crucial importance of introducing online social network site for marketing tourism businesses and services in Nigeria, and increasing the awareness for Nigeria as a tourist destination. Moreover, the paper strongly recommends the use of online social network as a tool for marketing tourism businesses and services, and the need for identifying effective framework for application of ICT tools in marketing tourism businesses and services in Nigeria at large.

Keywords: tourism business, internet, online social networks, tourism services, ICT

Procedia PDF Downloads 361

Authors: Baladitya Swaika, Rahul Khatry

Abstract:

Computer Adaptive Tests (CATs) is one of the most efficient ways for testing the cognitive abilities of students. CATs are based on Item Response Theory (IRT) which is based on item selection and ability estimation using statistical methods of maximum information selection/selection from posterior and maximum-likelihood (ML)/maximum a posteriori (MAP) estimators respectively. This study aims at combining both classical and Bayesian approaches to IRT to create a dataset which is then fed to a neural network which automates the process of ability estimation and then comparing it to traditional CAT models designed using IRT. This study uses python as the base coding language, pymc for statistical modelling of the IRT and scikit-learn for neural network implementations. On creation of the model and on comparison, it is found that the Neural Network based model performs 7-10% worse than the IRT model for score estimations. Although performing poorly, compared to the IRT model, the neural network model can be beneficially used in back-ends for reducing time complexity as the IRT model would have to re-calculate the ability every-time it gets a request whereas the prediction from a neural network could be done in a single step for an existing trained Regressor. This study also proposes a new kind of framework whereby the neural network model could be used to incorporate feature sets, other than the normal IRT feature set and use a neural network’s capacity of learning unknown functions to give rise to better CAT models. Categorical features like test type, etc. could be learnt and incorporated in IRT functions with the help of techniques like logistic regression and can be used to learn functions and expressed as models which may not be trivial to be expressed via equations. This kind of a framework, when implemented would be highly advantageous in psychometrics and cognitive assessments. This study gives a brief overview as to how neural networks can be used in adaptive testing, not only by reducing time-complexity but also by being able to incorporate newer and better datasets which would eventually lead to higher quality testing.

Keywords: computer adaptive tests, item response theory, machine learning, neural networks

Procedia PDF Downloads 180

Authors: Said Benaissa, David Plets, Emmeric Tanghe, Jens Trogh, Luc Martens, Leen Vandaele, Annelies Van Nuffel, Frank A. M. Tuyttens, Bart Sonck, Wout Joseph

Abstract:

The herd monitoring and managing - in particular the detection of ‘attention animals’ that require care, treatment or assistance is crucial for effective reproduction status, health, and overall well-being of dairy cows. In large sized farms, traditional methods based on direct observation or analysis of video recordings become labour-intensive and time-consuming. Thus, automatic monitoring systems using sensors have become increasingly important to continuously and accurately track the health status of dairy cows. Wireless sensor networks (WSNs) and internet-of-things (IoT) can be effectively used in health tracking of dairy cows to facilitate herd management and enhance the cow welfare. Since on-cow measuring devices are energy-constrained, a proper characterization of the off-body wireless channel between the on-cow sensor nodes and the back-end base station is required for a power-optimized deployment of these networks in barns. The aim of this study was to characterize the off-body wireless channel in indoor (barns) environment at 868 MHz using LoRa nodes. LoRa is an emerging wireless technology mainly targeted at WSNs and IoT networks. Both large scale fading (i.e., path loss) and temporal fading were investigated. The obtained path loss values as a function of the transmitter-receiver separation were well fitted by a lognormal path loss model. The path loss showed an additional increase of 4 dB when the wireless node was actually worn by the cow. The temporal fading due to movement of other cows was well described by Rician distributions with a K-factor of 8.5 dB. Based on this characterization, network planning and energy consumption optimization of the on-body wireless nodes could be performed, which enables the deployment of reliable dairy cow monitoring systems.

Keywords: channel, channel modelling, cow monitoring, dairy cows, health monitoring, IoT, LoRa, off-body propagation, PLF, propagation

Procedia PDF Downloads 322

Authors: Uma Maheshwari V., Rajanikanth Aluvalu, Kumar Gautam

Abstract:

The COVID-19 disease is a highly contagious viral infection with major worldwide health implications. The global economy suffers as a result of COVID. The spread of this pandemic disease can be slowed if positive patients are found early. COVID-19 disease prediction is beneficial for identifying patients' health problems that are at risk for COVID. Deep learning and machine learning algorithms for COVID prediction using X-rays have the potential to be extremely useful in solving the scarcity of doctors and clinicians in remote places. In this paper, a convolutional neural network (CNN) with deep layers is presented for recognizing COVID-19 patients using real-world datasets. We gathered around 6000 X-ray scan images from various sources and split them into two categories: normal and COVID-impacted. Our model examines chest X-ray images to recognize such patients. Because X-rays are commonly available and affordable, our findings show that X-ray analysis is effective in COVID diagnosis. The predictions performed well, with an average accuracy of 99% on training photographs and 88% on X-ray test images.

Keywords: deep CNN, COVID–19 analysis, feature extraction, feature map, accuracy

Procedia PDF Downloads 84

Authors: Emilio Galdeano-Gómez, Juan C. Pérez-Mesa, Laura Piedra-Muñoz, María C. García-Barranco, Jesús Hernández-Rubio

Abstract:

The situation in which firms engage in simultaneous cooperation and competition with each other is a phenomenon known as co-opetition. This scenario has received increasing attention in business economics and management analyses. In the domain of supply chain networks and for small and medium-sized enterprises, SMEs, these strategies are of greater relevance given the complex environment of globalization and competition in open markets. These firms face greater challenges regarding technology and access to specific resources due to their limited capabilities and limited market presence. Consequently, alliances and collaborations with both buyers and suppliers prove to be key elements in overcoming these constraints. However, rivalry and competition are also regarded as major factors in successful internationalization processes, as they are drivers for firms to attain a greater degree of specialization and to improve efficiency, for example enabling them to allocate scarce resources optimally and providing incentives for innovation and entrepreneurship. The present work aims to contribute to the literature on SMEs’ internationalization strategies. The sample is constituted by a panel data of marketing firms from the Andalusian food sector and a multivariate regression analysis is developed, measuring variables of co-opetition and international activity. The hierarchical regression equations method has been followed, thus resulting in three estimated models: the first one excluding the variables indicative of channel type, while the latter two include the international retailer chain and wholesaler variable. The findings show that the combination of several factors leads to a complex scenario of inter-organizational relationships of cooperation and competition. In supply chain management analyses, these relationships tend to be classified as either buyer-supplier (vertical level) or supplier-supplier relationships (horizontal level). Several buyers and suppliers tend to participate in supply chain networks, and in which the form of governance (hierarchical and non-hierarchical) influences cooperation and competition strategies. For instance, due to their market power and/or their closeness to the end consumer, some buyers (e.g. large retailers in food markets) can exert an influence on the selection and interaction of several of their intermediate suppliers, thus endowing certain networks in the supply chain with greater stability. This hierarchical influence may in turn allow these suppliers to develop their capabilities (e.g. specialization) to a greater extent. On the other hand, for those suppliers that are outside these networks, this environment of hierarchy, characterized by a “hub firm” or “channel master”, may provide an incentive for developing their co-opetition relationships. These results prove that the analyzed firms have experienced considerable growth in sales to new foreign markets, mainly in Europe, dealing with large retail chains and wholesalers as main buyers. This supply industry is predominantly made up of numerous SMEs, which has implied a certain disadvantage when dealing with the buyers, as negotiations have traditionally been held on an individual basis and in the face of high competition among suppliers. Over recent years, however, cooperation among these marketing firms has become more common, for example regarding R&D, promotion, scheduling of production and sales.

Keywords: co-petition networks, international supply chain, maketing agrifood firms, SMEs strategies

Procedia PDF Downloads 84

Authors: Mazen Al-Bulaihed, Nissar Wani, Abdulrahman Al-Arainy, Yasin Khan

Abstract:

Partial Discharge (PD) measurements are widely used for diagnostic purposes in electrical equipment used in power systems. The main cause of these measurements is to prevent large power failures as cables are prone to aging, which usually results in embrittlement, cracking and eventual failure of the insulating and sheathing materials, exposing the conductor and risking a potential short circuit, a likely cause of the electrical fire. Many distribution networks rely heavily on medium voltage (MV) power cables. The presence of joints in these networks is a vital part of serving the consumer demand for electricity continuously. Such measurements become even more important when the extent of dependence increases. Moreover, it is known that the partial discharge in joints and termination are difficult to track and are the most crucial point of failures in large power systems. This paper discusses the diagnostic techniques of four samples of XLPE insulated cable joints, each included with a different type of defect. Experiments were carried out by measuring PD and tanδ at very low frequency applied high voltage. The results show the importance of combining PD and tanδ for effective cable assessment.

Keywords: partial discharge, tan delta, very low frequency, XLPE cable

Procedia PDF Downloads 166

Authors: Chothmal

Abstract:

In this paper, we present a route discovery process which uses the signal strength on a link as a parameter of its inclusion in the route discovery method. The proposed signal-to-interference and noise ratio (SINR) based multipath reactive routing protocol is named as SINR-MP protocol. The proposed SINR-MP routing protocols has two following two features: a) SINR-MP protocol selects routes based on the SINR of the links during the route discovery process therefore it select the routes which has long lifetime and low frame error rate for data transmission, and b) SINR-MP protocols route discovery process is multipath which discovers more than one SINR based route between a given source destination pair. The multiple routes selected by our SINR-MP protocol are node-disjoint in nature which increases their robustness against link failures, as failure of one route will not affect the other route. The secondary route is very useful in situations where the primary route is broken because we can now use the secondary route without causing a new route discovery process. Due to this, the network overhead caused by a route discovery process is avoided. This increases the network performance greatly. The proposed SINR-MP routing protocol is implemented in the trail version of network simulator called Qualnet.

Keywords: ad hoc networks, quality of service, video streaming, H.264/SVC, multiple routes, video traces

Procedia PDF Downloads 254

Authors: Derkaoui Orkia, Lehireche Ahmed

Abstract:

The topic of this article is to exploring the potentialities of a powerful optimization technique, namely Semidefinite Programming, for solving NP-hard problems. This approach provides tight relaxations of combinatorial and quadratic problems. In this work, we solve the maximum clique problem using this relaxation. The clique problem is the computational problem of finding cliques in a graph. It is widely acknowledged for its many applications in real-world problems. The numerical results show that it is possible to find a maximum clique in polynomial time, using an algorithm based on semidefinite programming. We implement a primal-dual interior points algorithm to solve this problem based on semidefinite programming. The semidefinite relaxation of this problem can be solved in polynomial time.

Keywords: semidefinite programming, maximum clique problem, primal-dual interior point method, relaxation

Procedia PDF Downloads 224

Authors: C. P. Autry, A. W. Roscoe, Mykhailo Magal

Abstract:

We discuss the theoretical model underlying 2BPA (two-band peer authentication), a practical alternative to conventional authentication of entities and data in IoT. In essence, this involves assembling a virtual map of authentication assets in the network, typically leading to many paths of confirmation between any pair of entities. This map is continuously updated, confirmed, and evaluated. The value of authentication along multiple disjoint paths becomes very clear, and we require analogues of triangulation to extend authentication along extended paths and deliver it along all possible paths. We discover that if an attacker wants to make an honest node falsely believe she has authenticated another, then the length of the authentication paths is of little importance. This is because optimal attack strategies correspond to minimal cuts in the authentication graph and do not contain multiple edges on the same path. The authentication provided by disjoint paths normally is additive (in entropy).

Keywords: authentication, edge computing, industrial IoT, post-quantum resistance

Procedia PDF Downloads 204

Authors: Ali A. Hammadi

Abstract:

In this work, a practical study on a commissioned Giga Passive Optical Network (GPON) fiber to the home access network in Kuwait is presented. The work covers the framework of the conceptual design of the deployed Passive Optical Networks (PONs), access network, optical fiber cable network distribution, technologies, and standards. The work also describes methodologies applied by system engineers for design of Optical Network Terminals (ONTs) and Optical Line Terminals (OLTs) transceivers with respect to the distance, operating wavelengths, splitting ratios. The results have demonstrated and justified the limitation of transmission distance of a PON link in Fiber to The Premises (FTTP) to not exceed 20 km. Optical Time Domain Reflector (OTDR) test has been carried for this project to confirm compliance with International Telecommunication Union (ITU) specifications regarding the total length of the deployed optical cable, total loss in dB, and loss per km in dB/km with respect to the operating wavelengths. OTDR test results with traces for segments of implemented fiber network will be provided and discussed.

Keywords: passive optical networks (PONs), fiber to the premises (FTTx), access network, OTDR

Procedia PDF Downloads 292

Authors: Mikheil Kalmakhelidze

Abstract:

Record linkage (RL) problem has become more and more important in recent years due to the growing interest towards big data analysis. The problem can be formulated in a very simple way: Given two entries a and b of a database, decide whether they represent the same object or not. There are two classical deterministic and probabilistic ways of solving the RL problem. Using simple Bayes classifier in many cases produces useful results but sometimes they show to be poor. In recent years several successful approaches have been made towards solving specific RL problems by neural network algorithms including single layer perception, multilayer back propagation network etc. In our work, we model the RL problem for specific dataset of student applications in fuzzy description logic (FDL) where linkage of specific pair (a,b) depends on the truth value of corresponding formula A(a,b) in a canonical FDL model. As a main result, we build neural network for deciding truth value of FDL formulas in a canonical model and thus link RL problem to machine learning. We apply the approach to dataset with 10000 entries and also compare to classical RL solving approaches. The results show to be more accurate than standard probabilistic approach.

Keywords: description logic, fuzzy logic, neural networks, record linkage

Procedia PDF Downloads 275

Authors: C. S. Subhashini, H. L. Premaratne

Abstract:

Landslides are the most recurrent and prominent disaster in Sri Lanka. Sri Lanka has been subjected to a number of extreme landslide disasters that resulted in a significant loss of life, material damage, and distress. It is required to explore a solution towards preparedness and mitigation to reduce recurrent losses associated with landslides. Artificial Neural Networks (ANNs) and Hidden Markov Model (HMMs) are now widely used in many computer applications spanning multiple domains. This research examines the effectiveness of using Artificial Neural Networks and Hidden Markov Model in landslides predictions and the possibility of applying the modern technology to predict landslides in a prominent geographical area in Sri Lanka. A thorough survey was conducted with the participation of resource persons from several national universities in Sri Lanka to identify and rank the influencing factors for landslides. A landslide database was created using existing topographic; soil, drainage, land cover maps and historical data. The landslide related factors which include external factors (Rainfall and Number of Previous Occurrences) and internal factors (Soil Material, Geology, Land Use, Curvature, Soil Texture, Slope, Aspect, Soil Drainage, and Soil Effective Thickness) are extracted from the landslide database. These factors are used to recognize the possibility to occur landslides by using an ANN and HMM. The model acquires the relationship between the factors of landslide and its hazard index during the training session. These models with landslide related factors as the inputs will be trained to predict three classes namely, ‘landslide occurs’, ‘landslide does not occur’ and ‘landslide likely to occur’. Once trained, the models will be able to predict the most likely class for the prevailing data. Finally compared two models with regards to prediction accuracy, False Acceptance Rates and False Rejection rates and This research indicates that the Artificial Neural Network could be used as a strong decision support system to predict landslides efficiently and effectively than Hidden Markov Model.

Keywords: landslides, influencing factors, neural network model, hidden markov model

Procedia PDF Downloads 387

Authors: Y. Wang

Abstract:

The Traveling salesman problem (TSP) is NP-hard in combinatorial optimization. The research shows the algorithms for TSP on the sparse graphs have the shorter computation time than those for TSP according to the complete graphs. We present an improved iterative algorithm to compute the sparse graphs for TSP by frequency graphs computed with frequency quadrilaterals. The iterative algorithm is enhanced by adjusting two parameters of the algorithm. The computation time of the algorithm is O(CN_maxn²) where C is the iterations, N_max is the maximum number of frequency quadrilaterals containing each edge and n is the scale of TSP. The experimental results showed the computed sparse graphs generally have less than 5n edges for most of these Euclidean instances. Moreover, the maximum degree and minimum degree of the vertices in the sparse graphs do not have much difference. Thus, the computation time of the methods to resolve the TSP on these sparse graphs will be greatly reduced.

Keywords: frequency quadrilateral, iterative algorithm, sparse graph, traveling salesman problem

Procedia PDF Downloads 237

Authors: Xiaoxiao Li, Shuangcheng Jia, Qian Li

Abstract:

Keypoint detection has always been a challenge in the field of image recognition. This paper proposes a novelty keypoint detection method which is called Multi-Scale Feature Fusion Convolutional Network with Attention (MFFCNA). We verified that the multi-scale features with the attention mechanism module have better feature expression capability. The feature fusion between different scales makes the information that the network model can express more abundant, and the network is easier to converge. On our self-made street sign corner dataset, we validate the MFFCNA model with an accuracy of 97.8% and a recall of 81%, which are 5 and 8 percentage points higher than the HRNet network, respectively. On the COCO dataset, the AP is 71.9%, and the AR is 75.3%, which are 3 points and 2 points higher than HRNet, respectively. Extensive experiments show that our method has a remarkable improvement in the keypoint recognition tasks, and the recognition effect is better than the existing methods. Moreover, our method can be applied not only to keypoint detection but also to image classification and semantic segmentation with good generality.

Keywords: keypoint detection, feature fusion, attention, semantic segmentation

Procedia PDF Downloads 123

Authors: Andrew N. Saylor, James R. Peters

Abstract:

Scoliosis is a complex 3D deformity of the thoracic and lumbar spines, clinically diagnosed by measurement of a Cobb angle of 10 degrees or more on a coronal X-ray. The Cobb angle is the angle made by the lines drawn along the proximal and distal endplates of the respective proximal and distal vertebrae comprising the curve. Traditionally, Cobb angles are measured manually using either a marker, straight edge, and protractor or image measurement software. The task of measuring the Cobb angle can also be represented by a function taking the spine geometry rendered using X-ray imaging as input and returning the approximate angle. Although the form of such a function may be unknown, it can be approximated using artificial neural networks (ANNs). The performance of ANNs is affected by many factors, including the choice of activation function and network architecture; however, the effects of these parameters on the accuracy of scoliotic deformity measurements are poorly understood. Therefore, the objective of this study was to systematically investigate the effect of ANN architecture and activation function on Cobb angle measurement from the coronal X-rays of scoliotic subjects. The data set for this study consisted of 609 coronal chest X-rays of scoliotic subjects divided into 481 training images and 128 test images. These data, which included labeled Cobb angle measurements, were obtained from the SpineWeb online database. In order to normalize the input data, each image was resized using bi-linear interpolation to a size of 500 × 187 pixels, and the pixel intensities were scaled to be between 0 and 1. A fully connected (dense) ANN with a fixed cost function (mean squared error), batch size (10), and learning rate (0.01) was developed using Python Version 3.7.3 and TensorFlow 1.13.1. The activation functions (sigmoid, hyperbolic tangent [tanh], or rectified linear units [ReLU]), number of hidden layers (1, 3, 5, or 10), and number of neurons per layer (10, 100, or 1000) were varied systematically to generate a total of 36 network conditions. Stochastic gradient descent with early stopping was used to train each network. Three trials were run per condition, and the final mean squared errors and mean absolute errors were averaged to quantify the network response for each condition. The network that performed the best used ReLU neurons had three hidden layers, and 100 neurons per layer. The average mean squared error of this network was 222.28 ± 30 degrees2, and the average mean absolute error was 11.96 ± 0.64 degrees. It is also notable that while most of the networks performed similarly, the networks using ReLU neurons, 10 hidden layers, and 1000 neurons per layer, and those using Tanh neurons, one hidden layer, and 10 neurons per layer performed markedly worse with average mean squared errors greater than 400 degrees2 and average mean absolute errors greater than 16 degrees. From the results of this study, it can be seen that the choice of ANN architecture and activation function has a clear impact on Cobb angle inference from coronal X-rays of scoliotic subjects.

Keywords: scoliosis, artificial neural networks, cobb angle, medical imaging

Procedia PDF Downloads 136

Authors: Thibaud Berthomier, Ali Mansour, Luc Bressollette, Frédéric Le Roy, Dominique Mottier, Léo Fréchier, Barthélémy Hermenault

Abstract:

Deep Venous Thrombosis (DVT) occurs when a thrombus is formed within a deep vein (most often in the legs). This disease can be deadly if a part or the whole thrombus reaches the lung and causes a Pulmonary Embolism (PE). This disorder, often asymptomatic, has multifactorial causes: immobilization, surgery, pregnancy, age, cancers, and genetic variations. Our project aims to relate the thrombus epidemiology (origins, patient predispositions, PE) to its structure using ultrasound images. Ultrasonography and elastography were collected using Toshiba Aplio 500 at Brest Hospital. This manuscript compares two classification approaches: spectral clustering and scattering operator. The former is based on the graph and matrix theories while the latter cascades wavelet convolutions with nonlinear modulus and averaging operators.

Keywords: deep venous thrombosis, ultrasonography, elastography, scattering operator, wavelet, spectral clustering

Procedia PDF Downloads 483

Authors: Hossein Aghababaei, Sergio Vitale, Giampaolo Ferraioli

Abstract:

In the context of Synthetic Aperture Radar (SAR) data, polarization is an important source of information for Earth's surface monitoring. SAR Systems are often considered to transmit only one polarization. This constraint leads to either single or dual polarimetric SAR imaging modalities. Single polarimetric systems operate with a fixed single polarization of both transmitted and received electromagnetic (EM) waves, resulting in a single acquisition channel. Dual polarimetric systems, on the other hand, transmit in one fixed polarization and receive in two orthogonal polarizations, resulting in two acquisition channels. Dual polarimetric systems are obviously more informative than single polarimetric systems and are increasingly being used for a variety of remote sensing applications. In dual polarimetric systems, the choice of polarizations for the transmitter and the receiver is open. The choice of circular transmit polarization and coherent dual linear receive polarizations forms a special dual polarimetric system called hybrid polarimetry, which brings the properties of rotational invariance to geometrical orientations of features in the scene and optimizes the design of the radar in terms of reliability, mass, and power constraints. The complete characterization of target scattering, however, requires fully polarimetric data, which can be acquired with systems that transmit two orthogonal polarizations. This adds further complexity to data acquisition and shortens the coverage area or swath of fully polarimetric images compared to the swath of dual or hybrid polarimetric images. The search for solutions to augment dual polarimetric data to full polarimetric data will therefore take advantage of full characterization and exploitation of the backscattered field over a wider coverage with less system complexity. Several methods for reconstructing fully polarimetric images using hybrid polarimetric data can be found in the literature. Although the improvements achieved by the newly investigated and experimented reconstruction techniques are undeniable, the existing methods are, however, mostly based upon model assumptions (especially the assumption of reflectance symmetry), which may limit their reliability and applicability to vegetation and forest scenarios. To overcome the problems of these techniques, this paper proposes a new framework for reconstructing fully polarimetric information from hybrid polarimetric data. The framework uses Deep Learning solutions to augment hybrid polarimetric data without relying on model assumptions. A convolutional neural network (CNN) with a specific architecture and loss function is defined for this augmentation problem by focusing on different scattering properties of the polarimetric data. In particular, the method controls the CNN training process with respect to several characteristic features of polarimetric images defined by the combination of different terms in the cost or loss function. The proposed method is experimentally validated with real data sets and compared with a well-known and standard approach from the literature. From the experiments, the reconstruction performance of the proposed framework is superior to conventional reconstruction methods. The pseudo fully polarimetric data reconstructed by the proposed method also agree well with the actual fully polarimetric images acquired by radar systems, confirming the reliability and efficiency of the proposed method.

Keywords: SAR image, polarimetric SAR image, convolutional neural network, deep learnig, deep neural network

Procedia PDF Downloads 74

Authors: Ahmed Alutaibi, Ganti Sudhakar

Abstract:

Providing better Quality-of-Service (QoS) to end users has been a challenging problem for researchers and service providers. Building applications relying on best effort network protocols hindered the adoption of guaranteed service parameters and, ultimately, Quality of Service. The introduction of Software Defined Networking (SDN) opened the door for a new paradigm shift towards a more controlled programmable configurable behavior. Openflow has been and still is the main implementation of the SDN vision. To facilitate better QoS for applications, the network must calculate and measure certain parameters. One of those parameters is the delay between the two ends of the connection. Using the power of SDN and the knowledge of application and network behavior, SDN networks can adjust to different conditions and specifications. In this paper, we use the capabilities of SDN to implement multiple algorithms to measure delay end-to-end not only inside the SDN network. The results of applying the algorithms on an emulated environment show that we can get measurements close to the emulated delay. The results also show that depending on the algorithm, load on the network and controller can differ. In addition, the transport layer handshake algorithm performs best among the tested algorithms. Out of the results and implementation, we show the limitations of Openflow and develop suggestions to solve them.

Keywords: software defined networking, quality of service, delay measurement, openflow, mininet

Procedia PDF Downloads 169

Authors: Ashraf Aly

Abstract:

The aim of this study is to examine the impact of 5G communication technology radiation on human health and airport safety. The term 5G refers to the fifth generation of wireless mobile technology. The 5G wireless technology will increase the number of high-frequency-powered base stations and other devices and browsing and download speeds, as well as improve the network connectivity and play a big part in improving the performance of integrated applications, such as self-driving cars, medical devices, and robotics. 4G was the latest embedded version of mobile networking technology called 4G, and 5G is the new version of wireless technology. 5G networks have more features than 4G networks, such as lower latency, higher capacity, and increased bandwidth compared to 4G. 5G network improvements over 4G will have big impacts on how people live, business, and work all over the world. But neither 4G nor 5G have been tested for safety and show harmful effects from this wireless radiation. This paper presents biological factors on the effects of 5G radiation on human health. 5G services use C-band radio frequencies; these frequencies are close to those used by radio altimeters, which represent important equipment for airport and aircraft safety. The aviation industry, telecommunications companies, and their regulators have been discussing and weighing these interference concerns for years.

Keywords: wireless communication, radiofrequency, Electromagnetic field, environmental issues

Procedia PDF Downloads 70

Authors: Mark Pizzato

Abstract:

How do ideas, images, and emotions in stage-plays and videos affect us? Do they evoke a greater awareness (or cognitive reappraisal of emotions) through possible shifts between left-cortical, right-cortical, and subcortical networks? To address these questions, this presentation summarizes the research of various neuroscientists, especially Bernard Baars and others involved in Global Workspace Theory, Matthew Lieberman in social neuroscience, Iain McGilchrist on left and right cortical functions, and Jaak Panksepp on the subcortical circuits of primal emotions. Through such research, this presentation offers an ‘inner theatre’ model of the brain, regarding major hubs of neural networks and our animal ancestry. It also considers recent experiments, by Mario Beauregard, on the cognitive reappraisal of sad, erotic, and aversive film clips. Finally, it applies the inner-theatre model and related research to survey results of theatre students who read and then watched the ‘To be or not to be’ speech in 8 different video versions (from stage and screen productions) of William Shakespeare’s Hamlet. Findings show that students become aware of left-cortical, right-cortical, and subcortical brain functions—and shifts between them—through staging and movie-making choices in each of the different videos.

Keywords: cognitive reappraisal, Hamlet, neuroscience, Shakespeare, theatre

Procedia PDF Downloads 322

Authors: Shanxiong Chen, Xu Han, Xiaolong Wang, Hui Ma

Abstract:

Yi is an ethnic group mainly living in mainland China, with its own spoken and written language systems, after development of thousands of years. Ancient Yi is one of the six ancient languages in the world, which keeps a record of the history of the Yi people and offers documents valuable for research into human civilization. Recognition of the characters in ancient Yi helps to transform the documents into an electronic form, making their storage and spreading convenient. Due to historical and regional limitations, research on recognition of ancient characters is still inadequate. Thus, deep learning technology was applied to the recognition of such characters. Five models were developed on the basis of the four-layer convolutional neural network (CNN). Alpha-Beta divergence was taken as a penalty term to re-encode output neurons of the five models. Two fully connected layers fulfilled the compression of the features. Finally, at the softmax layer, the orthographic features of ancient Yi characters were re-evaluated, their probability distributions were obtained, and characters with features of the highest probability were recognized. Tests conducted show that the method has achieved higher precision compared with the traditional CNN model for handwriting recognition of the ancient Yi.

Keywords: recognition, CNN, Yi character, divergence

Procedia PDF Downloads 168

Authors: Naveed Ghani

Abstract:

With the arrival of higher speed communication links and mature application running over the internet, the requirement for reliable, efficient and robust network designs rising day by day. Multi-Protocol Label Switching technology (MPLS) Virtual Private Networks (VPNs) have committed to provide optimal network services. They are gaining popularity in industry day by day. Enterprise customers are moving to service providers that offer MPLS VPNs. The main reason for this shifting is the capability of MPLS VPN to provide built in security features and any-to-any connectivity. MPLS VPNs improved the network performance due to fast label switching as compare to traditional IP Forwarding but traffic classification and policing was still required on per hop basis to enhance the performance of real time traffic which is delay sensitive (particularly voice and video). QoS (Quality of service) is the most important factor to prioritize enterprise networks’ real time traffic such as voice and video. This thesis is focused on the study of QoS parameters (e.g. delay, jitter and MOS (Mean Opinion Score)) for the real time traffic over MPLS VPNs. DiffServ (Differentiated Services) QoS model will be used over MPLS VPN network to get end-to-end service quality.

Keywords: network, MPLS, VPN, DiffServ, MPLS VPN, DiffServ QoS, QoS Model, GNS2

Procedia PDF Downloads 429

Authors: Mehran Asadi, Afrand Agah

Abstract:

In data mining, partitioning is used as a fundamental tool for classification. With the help of partitioning, we study the structure of data, which allows us to envision decision rules, which can be applied to classification trees. In this research, we used online social network dataset and all of its attributes (e.g., Node features, labels, etc.) to determine what constitutes an above average chance of being a divergent node. We used the R statistical computing language to conduct the analyses in this report. The data were found on the UC Irvine Machine Learning Repository. This research introduces the basic concepts of classification in online social networks. In this work, we utilize overfitting and describe different approaches for evaluation and performance comparison of different classification methods. In classification, the main objective is to categorize different items and assign them into different groups based on their properties and similarities. In data mining, recursive partitioning is being utilized to probe the structure of a data set, which allow us to envision decision rules and apply them to classify data into several groups. Estimating densities is hard, especially in high dimensions, with limited data. Of course, we do not know the densities, but we could estimate them using classical techniques. First, we calculated the correlation matrix of the dataset to see if any predictors are highly correlated with one another. By calculating the correlation coefficients for the predictor variables, we see that density is strongly correlated with transitivity. We initialized a data frame to easily compare the quality of the result classification methods and utilized decision trees (with k-fold cross validation to prune the tree). The method performed on this dataset is decision trees. Decision tree is a non-parametric classification method, which uses a set of rules to predict that each observation belongs to the most commonly occurring class label of the training data. Our method aggregates many decision trees to create an optimized model that is not susceptible to overfitting. When using a decision tree, however, it is important to use cross-validation to prune the tree in order to narrow it down to the most important variables.

Keywords: online social networks, data mining, social cloud computing, interaction and collaboration

Procedia PDF Downloads 164

Authors: Hossein Aghababaei, Sergio Vitale, Giampaolo ferraioli

Abstract:

In the context of Synthetic Aperture Radar (SAR) data, polarization is an important source of information for Earth's surface monitoring . SAR Systems are often considered to transmit only one polarization. This constraint leads to either single or dual polarimetric SAR imaging modalities. Single polarimetric systems operate with a fixed single polarization of both transmitted and received electromagnetic (EM) waves, resulting in a single acquisition channel. Dual polarimetric systems, on the other hand, transmit in one fixed polarization and receive in two orthogonal polarizations, resulting in two acquisition channels. Dual polarimetric systems are obviously more informative than single polarimetric systems and are increasingly being used for a variety of remote sensing applications. In dual polarimetric systems, the choice of polarizations for the transmitter and the receiver is open. The choice of circular transmit polarization and coherent dual linear receive polarizations forms a special dual polarimetric system called hybrid polarimetry, which brings the properties of rotational invariance to geometrical orientations of features in the scene and optimizes the design of the radar in terms of reliability, mass, and power constraints. The complete characterization of target scattering, however, requires fully polarimetric data, which can be acquired with systems that transmit two orthogonal polarizations. This adds further complexity to data acquisition and shortens the coverage area or swath of fully polarimetric images compared to the swath of dual or hybrid polarimetric images. The search for solutions to augment dual polarimetric data to full polarimetric data will therefore take advantage of full characterization and exploitation of the backscattered field over a wider coverage with less system complexity. Several methods for reconstructing fully polarimetric images using hybrid polarimetric data can be found in the literature. Although the improvements achieved by the newly investigated and experimented reconstruction techniques are undeniable, the existing methods are, however, mostly based upon model assumptions (especially the assumption of reflectance symmetry), which may limit their reliability and applicability to vegetation and forest scenarios. To overcome the problems of these techniques, this paper proposes a new framework for reconstructing fully polarimetric information from hybrid polarimetric data. The framework uses Deep Learning solutions to augment hybrid polarimetric data without relying on model assumptions. A convolutional neural network (CNN) with a specific architecture and loss function is defined for this augmentation problem by focusing on different scattering properties of the polarimetric data. In particular, the method controls the CNN training process with respect to several characteristic features of polarimetric images defined by the combination of different terms in the cost or loss function. The proposed method is experimentally validated with real data sets and compared with a well-known and standard approach from the literature. From the experiments, the reconstruction performance of the proposed framework is superior to conventional reconstruction methods. The pseudo fully polarimetric data reconstructed by the proposed method also agree well with the actual fully polarimetric images acquired by radar systems, confirming the reliability and efficiency of the proposed method.

Keywords: SAR image, deep learning, convolutional neural network, deep neural network, SAR polarimetry

Procedia PDF Downloads 98

Authors: Ved Kulkarni, Karthik Kini

Abstract:

This experiment aims to understand how the media affects the power markets in the mainland United States and study the duration of reaction time between news updates and actual price movements. it have taken into account electric utility companies trading in the NYSE and excluded companies that are more politically involved and move with higher sensitivity to Politics. The scrapper checks for any news related to keywords, which are predefined and stored for each specific company. Based on this, the classifier will allocate the effect into five categories: positive, negative, highly optimistic, highly negative, or neutral. The effect on the respective price movement will be studied to understand the response time. Based on the response time observed, neural networks would be trained to understand and react to changing market conditions, achieving the best strategy in every market. The stock trader would be day trading in the first phase and making option strategy predictions based on the black holes model. The expected result is to create an AI-based system that adjusts trading strategies within the market response time to each price movement.

Keywords: data mining, language processing, artificial neural networks, sentiment analysis

Procedia PDF Downloads 24

Authors: Amirhossein Mohajerzadeh, Abolghasem Mohajerzadeh

Abstract:

In this paper, target parameter is estimated with desirable precision in hierarchical wireless sensor networks (WSN) while the proposed algorithm also tries to prolong network lifetime as much as possible, using efficient data collecting algorithm. Target parameter distribution function is considered unknown. Sensor nodes sense the environment and send the data to the base station called fusion center (FC) using hierarchical data collecting algorithm. FC builds underlying phenomena based on collected data. Considering the aggregation level, x, the goal is providing the essential infrastructure to find the best value for aggregation level in order to prolong network lifetime as much as possible, while desirable accuracy is guaranteed (required sample size is fully depended on desirable precision). First, the sample size calculation algorithm is discussed, second, the average queue length based on M/M[x]/1/K queue model is determined and it is used for energy consumption calculation. Nodes can decrease transmission cost by aggregating incoming data. Furthermore, the performance of the new algorithm is evaluated in terms of lifetime and estimation accuracy.

Keywords: aggregation, estimation, queuing, wireless sensor network

Procedia PDF Downloads 189