Search results for: meta-heuristic algorithms

Authors: Yehjune Heo

Abstract:

As biometric systems become widely deployed, the security of identification systems can be easily attacked by various spoof materials. This paper contributes to finding a reliable and practical anti-spoofing method using Convolutional Neural Networks (CNNs) based on the types of loss functions and optimizers. The types of CNNs used in this paper include AlexNet, VGGNet, and ResNet. By using various loss functions including Cross-Entropy, Center Loss, Cosine Proximity, and Hinge Loss, and various loss optimizers which include Adam, SGD, RMSProp, Adadelta, Adagrad, and Nadam, we obtained significant performance changes. We realize that choosing the correct loss function for each model is crucial since different loss functions lead to different errors on the same evaluation. By using a subset of the Livdet 2017 database, we validate our approach to compare the generalization power. It is important to note that we use a subset of LiveDet and the database is the same across all training and testing for each model. This way, we can compare the performance, in terms of generalization, for the unseen data across all different models. The best CNN (AlexNet) with the appropriate loss function and optimizers result in more than 3% of performance gain over the other CNN models with the default loss function and optimizer. In addition to the highest generalization performance, this paper also contains the models with high accuracy associated with parameters and mean average error rates to find the model that consumes the least memory and computation time for training and testing. Although AlexNet has less complexity over other CNN models, it is proven to be very efficient. For practical anti-spoofing systems, the deployed version should use a small amount of memory and should run very fast with high anti-spoofing performance. For our deployed version on smartphones, additional processing steps, such as quantization and pruning algorithms, have been applied in our final model.

Keywords: anti-spoofing, CNN, fingerprint recognition, loss function, optimizer

Procedia PDF Downloads 110

Authors: Ainouna Bouziane

Abstract:

The ability of Electron Tomography to recover the 3D structure of catalysts, with spatial resolution in the subnanometer scale, has been widely explored and reviewed in the last decades. A variety of experimental techniques, based either on Transmission Electron Microscopy (TEM) or Scanning Transmission Electron Microscopy (STEM) have been used to reveal different features of nanostructured catalysts in 3D, but High Angle Annular Dark Field imaging in STEM mode (HAADF-STEM) stands out as the most frequently used, given its chemical sensitivity and avoidance of imaging artifacts related to diffraction phenomena when dealing with crystalline materials. In this regard, our group has developed a methodology that combines image denoising by undecimated wavelet transforms (UWT) with automated, advanced segmentation procedures and parameter selection methods using CS-TVM (Compressed Sensing-total variation minimization) algorithms to reveal more reliable quantitative information out of the 3D characterization studies. However, evaluating the accuracy of the magnitudes estimated from the segmented volumes is also an important issue that has not been properly addressed yet, because a perfectly known reference is needed. The problem particularly complicates in the case of multicomponent material systems. To tackle this key question, we have developed a methodology that incorporates volume reconstruction/segmentation methods. In particular, we have established an approach to evaluate, in quantitative terms, the accuracy of TVM reconstructions, which considers the influence of relevant experimental parameters like the range of tilt angles, image noise level or object orientation. The approach is based on the analysis of material-realistic, 3D phantoms, which include the most relevant features of the system under analysis.

Keywords: electron tomography, supported catalysts, nanometrology, error assessment

Procedia PDF Downloads 58

Authors: Manju Kanu, Subrata Sinha, Surabhi Johari

Abstract:

Viral proteins of Ebola viruses belong to one of the best studied viruses; however no effective prevention against EBOV has been developed. Epitope-based vaccines provide a new strategy for prophylactic and therapeutic application of pathogen-specific immunity. A critical requirement of this strategy is the identification and selection of T-cell epitopes that act as vaccine targets. This study describes current methodologies for the selection process, with Ebola virus as a model system. Hence great challenge in the field of ebola virus research is to design universal vaccine. A combination of publicly available bioinformatics algorithms and computational tools are used to screen and select antigen sequences as potential T-cell epitopes of supertypes Human Leukocyte Antigen (HLA) alleles. MUSCLE and MOTIF tools were used to find out most conserved peptide sequences of viral proteins. Immunoinformatics tools were used for prediction of immunogenic peptides of viral proteins in zaire strains of Ebola virus. Putative epitopes for viral proteins (VP) were predicted from conserved peptide sequences of VP. Three tools NetCTL 1.2, BIMAS and Syfpeithi were used to predict the Class I putative epitopes while three tools, ProPred, IEDB-SMM-align and NetMHCII 2.2 were used to predict the Class II putative epitopes. B cell epitopes were predicted by BCPREDS 1.0. Immunogenic peptides were identified and selected manually by putative epitopes predicted from online tools individually for both MHC classes. Finally sequences of predicted peptides for both MHC classes were looked for common region which was selected as common immunogenic peptide. The immunogenic peptides were found for viral proteins of Ebola virus: epitopes FLESGAVKY, SSLAKHGEY. These predicted peptides could be promising candidates to be used as target for vaccine design.

Keywords: epitope, b cell, immunogenicity, ebola

Procedia PDF Downloads 286

Authors: Newton Muhury, Armando A. Apan, Tek N. Marasani, Gebiaw T. Ayele

Abstract:

The changing climate has degraded freshwater availability in Australia that influencing vegetation growth to a great extent. This study assessed the vegetation responses to groundwater using Terra’s moderate resolution imaging spectroradiometer (MODIS), Normalised Difference Vegetation Index (NDVI), and soil water content (SWC). A hydrological model, SWAT, has been set up in a southeast Australian river catchment for groundwater analysis. The model was calibrated and validated against monthly streamflow from 2001 to 2006 and 2007 to 2010, respectively. The SWAT simulated soil water content for 43 sub-basins and monthly MODIS NDVI data for three different types of vegetation (forest, shrub, and grass) were applied in the machine learning tool, Waikato Environment for Knowledge Analysis (WEKA), using two supervised machine learning algorithms, i.e., support vector machine (SVM) and random forest (RF). The assessment shows that different types of vegetation response and soil water content vary in the dry and wet seasons. The WEKA model generated high positive relationships (r = 0.76, 0.73, and 0.81) between NDVI values of all vegetation in the sub-basins against soil water content (SWC), the groundwater flow (GW), and the combination of these two variables, respectively, during the dry season. However, these responses were reduced by 36.8% (r = 0.48) and 13.6% (r = 0.63) against GW and SWC, respectively, in the wet season. Although the rainfall pattern is highly variable in the study area, the summer rainfall is very effective for the growth of the grass vegetation type. This study has enriched our knowledge of vegetation responses to groundwater in each season, which will facilitate better floodplain vegetation management.

Keywords: ArcSWAT, machine learning, floodplain vegetation, MODIS NDVI, groundwater

Procedia PDF Downloads 71

Authors: Abdul Basit Kiani

Abstract:

The software for the digital library system is a cutting-edge solution designed to revolutionize the way libraries manage and provide access to their vast collections of digital content. This advanced software leverages the power of technology to offer a seamless and user-friendly experience for both library staff and patrons. By implementing this software, libraries can efficiently organize, store, and retrieve digital resources, including e-books, audiobooks, journals, articles, and multimedia content. Its intuitive interface allows library staff to effortlessly manage cataloging, metadata extraction, and content enrichment, ensuring accurate and comprehensive access to digital materials. For patrons, the software offers a personalized and immersive digital library experience. They can easily browse the digital catalog, search for specific items, and explore related content through intelligent recommendation algorithms. The software also facilitates seamless borrowing, lending, and preservation of digital items, enabling users to access their favorite resources anytime, anywhere, on multiple devices. With robust security features, the software ensures the protection of intellectual property rights and enforces access controls to safeguard sensitive content. Integration with external authentication systems and user management tools streamlines the library's administration processes, while advanced analytics provide valuable insights into patron behavior and content usage. Overall, this software for the digital library system empowers libraries to embrace the digital era, offering enhanced access, convenience, and discoverability of their vast collections. It paves the way for a more inclusive and engaging library experience, catering to the evolving needs of tech-savvy patrons.

Keywords: software development, empowering digital libraries, digital cataloging and access, management system

Procedia PDF Downloads 48

Authors: Thomas Schalow

Abstract:

This paper posits the need to take a cross-cultural approach to communication with non-human cultures and intelligences in order to meet the following three imminent contingencies: communicating with sentient biological intelligences, communicating with extraterrestrial intelligences, and communicating with artificial super-intelligences. The paper begins with a discussion of how intelligence emerges. It disputes some common assumptions we maintain about consciousness, intention, and language. The paper next explores cross-cultural communication among humans, including non-sapiens species. The next argument made is that we need to become much more serious about communicating with the non-human, intelligent life forms that already exist around us here on Earth. There is an urgent need to broaden our definition of communication and reach out to the other sentient life forms that inhabit our world. The paper next examines the science and philosophy behind CETI (communication with extraterrestrial intelligences) and how it has proven useful, even in the absence of contact with alien life. However, CETI’s assumptions and methodology need to be revised and based on the cross-cultural approach to communication proposed in this paper if we are truly serious about finding and communicating with life beyond Earth. The final theme explored in this paper is communication with non-biological super-intelligences using a cross-cultural communication approach. This will present a serious challenge for humanity, as we have never been truly compelled to converse with other species, and our failure to seriously consider such intercourse has left us largely unprepared to deal with communication in a future that will be mediated and controlled by computer algorithms. Fortunately, our experience dealing with other human cultures can provide us with a framework for this communication. The basic assumptions behind intercultural communication can be applied to the many types of communication envisioned in this paper if we are willing to recognize that we are in fact dealing with other cultures when we interact with other species, alien life, and artificial super-intelligence. The ideas considered in this paper will require a new mindset for humanity, but a new disposition will prepare us to face the challenges posed by a future dominated by artificial intelligence.

Keywords: artificial intelligence, CETI, communication, culture, language

Procedia PDF Downloads 329

Authors: F. Lazzeri, I. Reiter

Abstract:

Energy production optimization has been traditionally very important for utilities in order to improve resource consumption. However, load forecasting is a challenging task, as there are a large number of relevant variables that must be considered, and several strategies have been used to deal with this complex problem. This is especially true also in microgrids where many elements have to adjust their performance depending on the future generation and consumption conditions. The goal of this paper is to present a solution for short-term load forecasting in microgrids, based on three machine learning experiments developed in R and web services built and deployed with different components of Cortana Intelligence Suite: Azure Machine Learning, a fully managed cloud service that enables to easily build, deploy, and share predictive analytics solutions; SQL database, a Microsoft database service for app developers; and PowerBI, a suite of business analytics tools to analyze data and share insights. Our results show that Boosted Decision Tree and Fast Forest Quantile regression methods can be very useful to predict hourly short-term consumption in microgrids; moreover, we found that for these types of forecasting models, weather data (temperature, wind, humidity and dew point) can play a crucial role in improving the accuracy of the forecasting solution. Data cleaning and feature engineering methods performed in R and different types of machine learning algorithms (Boosted Decision Tree, Fast Forest Quantile and ARIMA) will be presented, and results and performance metrics discussed.

Keywords: time-series, features engineering methods for forecasting, energy demand forecasting, Azure Machine Learning

Procedia PDF Downloads 277

Authors: Rami Hashish, Manon Limousis-Gayda, Caitlin McCleery

Abstract:

Introduction: Mild traumatic brain injuries, also referred to as concussions, represent an increasing burden to society. Due to limited objective diagnostic measures, concussions are diagnosed by assessing subjective symptoms, often leading to disputes to their presence. Common biomechanical measures associated with concussion are high linear and/or angular acceleration to the head. With regards to linear acceleration, approximately 80g’s has previously been shown to equate with a 50% probability of concussion. Motor vehicle collisions (MVCs) are a leading cause of concussion, due to high head accelerations experienced. The change in velocity (delta-V) of a vehicle in an MVC is an established metric for impact severity. As acceleration is the rate of delta-V with respect to time, the purpose of this paper is to determine the relation between delta-V (and occupant parameters) with linear head acceleration. Methods: A meta-analysis was conducted for manuscripts collected using the following keywords: head acceleration, concussion, brain injury, head kinematics, delta-V, change in velocity, motor vehicle collision, and rear-end. Ultimately, 280 studies were surveyed, 14 of which fulfilled the inclusion criteria as studies investigating the human response to impacts, reporting head acceleration, and delta-V of the occupant’s vehicle. Statistical analysis was conducted with SPSS and R. The best fit line analysis allowed for an initial understanding of the relation between head acceleration and delta-V. To further investigate the effect of occupant parameters on head acceleration, a quadratic model and a full linear mixed model was developed. Results: From the 14 selected studies, 139 crashes were analyzed with head accelerations and delta-V values ranging from 0.6 to 17.2g and 1.3 to 11.1 km/h, respectively. Initial analysis indicated that the best line of fit (Model 1) was defined as Head Acceleration = 0.465

Keywords: acceleration, brain injury, change in velocity, Delta-V, TBI

Procedia PDF Downloads 207

Authors: Xiaoqing Wang, Jianjian Zong, Li Li, Yanxing Zheng, Jinrong Tong, Mao Zhan

Abstract:

With the rapid development of network and communication technology, a great deal of data has been generated in different domains of a network. These data show a trend of increasing scale and more complex structure. Therefore, an effective and flexible cross-domain data-sharing system is needed. The Cross-domain Data Sharing System(CDSS) in a net-centric environment is composed of three sub-systems. The data distribution sub-system provides data exchange service through publish-subscribe technology that supports asynchronism and multi-to-multi communication, which adapts to the needs of the dynamic and large-scale distributed computing environment. The access control sub-system adopts Attribute-Based Access Control(ABAC) technology to uniformly model various data attributes such as subject, object, permission and environment, which effectively monitors the activities of users accessing resources and ensures that legitimate users get effective access control rights within a legal time. The cross-domain access security negotiation subsystem automatically determines the access rights between different security domains in the process of interactive disclosure of digital certificates and access control policies through trust policy management and negotiation algorithms, which provides an effective means for cross-domain trust relationship establishment and access control in a distributed environment. The CDSS’s asynchronous,multi-to-multi and loosely-coupled communication features can adapt well to data exchange and sharing in dynamic, distributed and large-scale network environments. Next, we will give CDSS new features to support the mobile computing environment.

Keywords: data sharing, cross-domain, data exchange, publish-subscribe

Procedia PDF Downloads 103

Authors: Usha Barahmand, Ruhollah Heydari Sheikh Ahmad, Sara Alaie Khoraem

Abstract:

Introduction: The association of epilepsy with several psychological disorders and reduced quality of life has long been recognized. The present study aimed at comparing the personality profiles, quality of life and symptomatology of anxiety and depression in patients with epilepsy and healthy controls. Materials and Methods: Forty seven patients (29 men and 18 women) with diagnosed epilepsy participated in this study. Forty seven healthy controls who matched the patients in age and gender were also recruited. The participants’ personality and psychological profiles were assessed using the Depression, Anxiety, and Stress Scale (DASS-21), the Short-Form Health Survey (SF-36) and the HEXACO Personality Inventory (HEXACO-PI). Scoring algorithms were applied to the SF-36 produce the physical and mental component scores (PCS and MCS). Results: There were statistically significant differences in the total SF-36 score, anxiety, depression and stress scores of the DASS-21 between patients and controls. Anxiety, stress and depression scores significantly correlated inversely with the PCS and MCS. Data analysis showed that females had higher depression scores than males in both patients and controls, while males in both groups scored higher on stress. Patients’ personality scores were also different from those reported by controls on emotional, agreeableness and extroversion. Patients scored higher on emotionality, and lower on agreeableness and extraversion. Patients also scored lower on indices of quality of life. Regression analysis revealed that emotionality, anxiety, stress and MCS accounted for a significant proportion of the variance in severity of epileptic seizures. Conclusion: Stressful situations and psychological conditions as well as the personality trait of neuroticism were related to the occurrence of recurrent epileptic seizures.

Keywords: anxiety, depression, epilepsy, neuroticism, personality, quality of life, stress

Procedia PDF Downloads 343

Authors: Animish Jain

Abstract:

This study deals with using Multiplicative Weight Update within artificial intelligence and machine learning to create models that can diagnose skin cancer using microscopic images of cancer samples. In this study, the multiplicative weight update method is used to take the predictions of multiple models to try and acquire more accurate results. Logistic Regression, Convolutional Neural Network (CNN), and Support Vector Machine Classifier (SVMC) models are employed within the Multiplicative Weight Update system. These models are trained on pictures of skin cancer from the ISIC-Archive, to look for patterns to label unseen scans as either benign or malignant. These models are utilized in a multiplicative weight update algorithm which takes into account the precision and accuracy of each model through each successive guess to apply weights to their guess. These guesses and weights are then analyzed together to try and obtain the correct predictions. The research hypothesis for this study stated that there would be a significant difference in the accuracy of the three models and the Multiplicative Weight Update system. The SVMC model had an accuracy of 77.88%. The CNN model had an accuracy of 85.30%. The Logistic Regression model had an accuracy of 79.09%. Using Multiplicative Weight Update, the algorithm received an accuracy of 72.27%. The final conclusion that was drawn was that there was a significant difference in the accuracy of the three models and the Multiplicative Weight Update system. The conclusion was made that using a CNN model would be the best option for this problem rather than a Multiplicative Weight Update system. This is due to the possibility that Multiplicative Weight Update is not effective in a binary setting where there are only two possible classifications. In a categorical setting with multiple classes and groupings, a Multiplicative Weight Update system might become more proficient as it takes into account the strengths of multiple different models to classify images into multiple categories rather than only two categories, as shown in this study. This experimentation and computer science project can help to create better algorithms and models for the future of artificial intelligence in the medical imaging field.

Keywords: artificial intelligence, machine learning, multiplicative weight update, skin cancer

Procedia PDF Downloads 51

Authors: Luleka Dlamini, Olivier Crespo, Jos van Dam

Abstract:

Accurately estimating crop growth and yield is pivotal for timely sustainable agricultural management and ensuring food security. Crop models and remote sensing can complement each other and form a robust analysis tool to improve crop growth and yield estimations when combined. This study thus aims to systematically evaluate how research that exclusively focuses on assimilating RS data into crop models varies among countries, crops, data assimilation methods, and farming conditions. A strict search string was applied in the Scopus and Web of Science databases, and 497 potential publications were obtained. After screening for relevance with predefined inclusion/exclusion criteria, 123 publications were considered in the final review. Results indicate that over 81% of the studies were conducted in countries associated with high socio-economic and technological advancement, mainly China, the United States of America, France, Germany, and Italy. Many of these studies integrated MODIS or Landsat data into WOFOST to improve crop growth and yield estimation of staple crops at the field and regional scales. Most studies use recalibration or updating methods alongside various algorithms to assimilate remotely sensed leaf area index into crop models. However, these methods cannot account for the uncertainties in remote sensing observations and the crop model itself. l. Over 85% of the studies were based on commercial and irrigated farming systems. Despite a great global interest in data assimilation into crop models, limited research has been conducted in resource- and data-limited regions like Africa. We foresee a great potential for such application in those conditions. Hence facilitating and expanding the use of such an approach, from which developing farming communities could benefit.

Keywords: crop models, remote sensing, data assimilation, crop yield estimation

Procedia PDF Downloads 94

Authors: Luleka Dlamini, Olivier Crespo, Jos van Dam

Abstract:

Accurately estimating crop growth and yield is pivotal for timely sustainable agricultural management and ensuring food security. Crop models and remote sensing can complement each other and form a robust analysis tool to improve crop growth and yield estimations when combined. This study thus aims to systematically evaluate how research that exclusively focuses on assimilating RS data into crop models varies among countries, crops, data assimilation methods, and farming conditions. A strict search string was applied in the Scopus and Web of Science databases, and 497 potential publications were obtained. After screening for relevance with predefined inclusion/exclusion criteria, 123 publications were considered in the final review. Results indicate that over 81% of the studies were conducted in countries associated with high socio-economic and technological advancement, mainly China, the United States of America, France, Germany, and Italy. Many of these studies integrated MODIS or Landsat data into WOFOST to improve crop growth and yield estimation of staple crops at the field and regional scales. Most studies use recalibration or updating methods alongside various algorithms to assimilate remotely sensed leaf area index into crop models. However, these methods cannot account for the uncertainties in remote sensing observations and the crop model itself. l. Over 85% of the studies were based on commercial and irrigated farming systems. Despite a great global interest in data assimilation into crop models, limited research has been conducted in resource- and data-limited regions like Africa. We foresee a great potential for such application in those conditions. Hence facilitating and expanding the use of such an approach, from which developing farming communities could benefit.

Keywords: crop models, remote sensing, data assimilation, crop yield estimation

Procedia PDF Downloads 58

Authors: Antonio Horta, Renato Marinho, Raimir Holanda

Abstract:

Considered as an evolution of the Blockchain, the Ethereum platform, besides allowing transactions of its cryptocurrency named Ether, it allows the programming of decentralised applications (DApps) and smart contracts. However, this functionality into blockchains has raised other types of threats, and the exploitation of smart contracts vulnerabilities has taken companies to experience big losses. This research intends to figure out the number of contracts that are under risk of being drained. Through a deep investigation, more than two hundred thousand smart contracts currently available in the Ethereum platform were scanned and estimated how much money is at risk. The experiment was based in a query run on Google Big Query in July 2022 and returned 50,707,133 contracts published on the Ethereum platform. After applying the filtering criteria, the experimentgot 430,584 smart contracts to download and analyse. The filtering criteria consisted of filtering out: ERC20 and ERC721 contracts, contracts without transactions, and contracts without balance. From this amount of 430,584 smart contracts selected, only 268,103 had source codes published on Etherscan, however, we discovered, using a hashing process, that there were contracts duplication. Removing the duplicated contracts, the process ended up with 20,417 source codes, which were analysed using the open source SAST tool smartbugswith oyente and securify algorithms. In the end, there was nearly $100,000 at risk of being drained from the potentially vulnerable smart contracts. It is important to note that the tools used in this study may generate false positives, which may interfere with the number of vulnerable contracts. To address this point, our next step in this research is to develop an application to test the contract in a parallel environment to verify the vulnerability. Finally, this study aims to alert users and companies about the risk on not properly creating and analysing their smart contracts before publishing them into the platform. As any other application, smart contracts are at risk of having vulnerabilities which, in this case, may result in direct financial losses.

Keywords: blockchain, reentrancy, static application security testing, smart contracts

Procedia PDF Downloads 66

Authors: Abdesselem Dakhli, Wajdi Bellil, Chokri Ben Amar

Abstract:

DNA Barcode, a short mitochondrial DNA fragment, made up of three subunits; a phosphate group, sugar and nucleic bases (A, T, C, and G). They provide good sources of information needed to classify living species. Such intuition has been confirmed by many experimental results. Species classification with DNA Barcode sequences has been studied by several researchers. The classification problem assigns unknown species to known ones by analyzing their Barcode. This task has to be supported with reliable methods and algorithms. To analyze species regions or entire genomes, it becomes necessary to use similarity sequence methods. A large set of sequences can be simultaneously compared using Multiple Sequence Alignment which is known to be NP-complete. To make this type of analysis feasible, heuristics, like progressive alignment, have been developed. Another tool for similarity search against a database of sequences is BLAST, which outputs shorter regions of high similarity between a query sequence and matched sequences in the database. However, all these methods are still computationally very expensive and require significant computational infrastructure. Our goal is to build predictive models that are highly accurate and interpretable. This method permits to avoid the complex problem of form and structure in different classes of organisms. On empirical data and their classification performances are compared with other methods. Our system consists of three phases. The first is called transformation, which is composed of three steps; Electron-Ion Interaction Pseudopotential (EIIP) for the codification of DNA Barcodes, Fourier Transform and Power Spectrum Signal Processing. The second is called approximation, which is empowered by the use of Multi Llibrary Wavelet Neural Networks (MLWNN).The third is called the classification of DNA Barcodes, which is realized by applying the algorithm of hierarchical classification.

Keywords: DNA barcode, electron-ion interaction pseudopotential, Multi Library Wavelet Neural Networks (MLWNN)

Procedia PDF Downloads 292

Authors: Waqqas-ur-Rehman Butt, Martin Servin, Marion Pause

Abstract:

In recent years, object detection has gained much attention and very encouraging research area in the field of computer vision. The robust object boundaries detection in an image is demanded in numerous applications of human computer interaction and automated surveillance systems. Many methods and approaches have been developed for automatic object detection in various fields, such as automotive, quality control management and environmental services. Inappropriately, to the best of our knowledge, object detection under illumination with shadow consideration has not been well solved yet. Furthermore, this problem is also one of the major hurdles to keeping an object detection method from the practical applications. This paper presents an approach to automatic object detection in images under non-standardized environmental conditions. A key challenge is how to detect the object, particularly under uneven illumination conditions. Image capturing conditions the algorithms need to consider a variety of possible environmental factors as the colour information, lightening and shadows varies from image to image. Existing methods mostly failed to produce the appropriate result due to variation in colour information, lightening effects, threshold specifications, histogram dependencies and colour ranges. To overcome these limitations we propose an object detection algorithm, with pre-processing methods, to reduce the interference caused by shadow and illumination effects without fixed parameters. We use the Y CrCb colour model without any specific colour ranges and predefined threshold values. The segmented object regions are further classified using morphological operations (Erosion and Dilation) and contours. Proposed approach applied on a large image data set acquired under various environmental conditions for wood stack detection. Experiments show the promising result of the proposed approach in comparison with existing methods.

Keywords: image processing, illumination equalization, shadow filtering, object detection

Procedia PDF Downloads 194

Authors: Oluwaseun K. Oyebode, Josiah A. Adeyemo

Abstract:

Hydrological modelling plays a crucial role in the planning and management of water resources, most especially in water stressed regions where the need to effectively manage the available water resources is of critical importance. However, due to the complex, nonlinear and dynamic behaviour of hydro-climatic interactions, achieving reliable modelling of water resource systems and accurate projection of hydrological parameters are extremely challenging. Although a significant number of modelling techniques (process-based and data-driven) have been developed and adopted in that regard, the field of hydrological modelling is still considered as one that has sluggishly progressed over the past decades. This is majorly as a result of the identification of some degree of uncertainty in the methodologies and results of techniques adopted. In recent times, evolutionary computation (EC) techniques have been developed and introduced in response to the search for efficient and reliable means of providing accurate solutions to hydrological related problems. This paper presents a comprehensive review of the underlying principles, methodological needs and applications of a promising evolutionary computation modelling technique – genetic programming (GP). It examines the specific characteristics of the technique which makes it suitable to solving hydrological modelling problems. It discusses the opportunities inherent in the application of GP in water related-studies such as rainfall estimation, rainfall-runoff modelling, streamflow forecasting, sediment transport modelling, water quality modelling and groundwater modelling among others. Furthermore, the means by which such opportunities could be harnessed in the near future are discussed. In all, a case for total embracement of GP and its variants in hydrological modelling studies is made so as to put in place strategies that would translate into achieving meaningful progress as it relates to modelling of water resource systems, and also positively influence decision-making by relevant stakeholders.

Keywords: computational modelling, evolutionary algorithms, genetic programming, hydrological modelling

Procedia PDF Downloads 270

Authors: Beatriz Lafuente Alcázar, Yash Wani, Amit J. Nimunkar

Abstract:

Cardiovascular Diseases (CVDs) are the leading cause of death globally, and around 80% of sudden cardiac deaths are due to arrhythmias or irregular heartbeats. The majority of these pathologies are revealed by either short-term or long-term alterations in the electrocardiogram (ECG) morphology. The ECG is the main diagnostic tool in cardiology. It is a non-invasive, pain free procedure that measures the heart’s electrical activity and that allows the detecting of abnormal rhythms and underlying conditions. A cardiologist can diagnose a wide range of pathologies based on ECG’s form alterations, but the human interpretation is subjective and it is contingent to error. Moreover, ECG records can be quite prolonged in time, which can further complicate visual diagnosis, and deeply retard disease detection. In this context, deep learning methods have risen as a promising strategy to extract relevant features and eliminate individual subjectivity in ECG analysis. They facilitate the computation of large sets of data and can provide early and precise diagnoses. Therefore, the cardiology field is one of the areas that can most benefit from the implementation of deep learning algorithms. In the present study, a deep learning algorithm is trained following a novel approach, using a combination of different databases as the training set. The goal of the algorithm is to achieve the detection of R-peaks in ECG signals. Its performance is further evaluated in ECG signals with different origins and features to test the model’s ability to generalize its outcomes. Performance of the model for detection of R-peaks for clean and noisy ECGs is presented. The model is able to detect R-peaks in the presence of various types of noise, and when presented with data, it has not been trained. It is expected that this approach will increase the effectiveness and capacity of cardiologists to detect divergences in the normal cardiac activity of their patients.

Keywords: arrhythmia, deep learning, electrocardiogram, machine learning, R-peaks

Procedia PDF Downloads 149

Authors: Jeff Chak-Fu Wong, Tony Chun Yin Yip

Abstract:

The transition from secondary school to university seems exciting for many first-year students but can be more challenging than expected. Enabling instructors to know students’ learning habits and styles enhances their understanding of the students’ learning backgrounds, allows teachers to provide better support for their students, and has therefore high potential to improve teaching quality and learning, especially in any mathematics-related courses. The aim of this research is to collect students’ data using online surveys, to analyze students’ factors using learning analytics and educational data mining and to discover the characteristics of the students at risk of falling behind in their studies based on students’ previous academic backgrounds and collected data. In this paper, we use correlation-based distance methods and mutual information for measuring student factor relationships. We then develop a factor network using the Minimum Spanning Tree method and consider further study for analyzing the topological properties of these networks using social network analysis tools. Under the framework of mutual information, two graph-based feature filtering methods, i.e., unsupervised and supervised infinite feature selection algorithms, are used to analyze the results for students’ data to rank and select the appropriate subsets of features and yield effective results in identifying the factors affecting students at risk of failing. This discovered knowledge may help students as well as instructors enhance educational quality by finding out possible under-performers at the beginning of the first semester and applying more special attention to them in order to help in their learning process and improve their learning outcomes.

Keywords: students' academic performance, correlation-based distance method, social network analysis, feature selection, graph-based feature filtering method

Procedia PDF Downloads 101

Authors: Michael Moeti, Khuliso Sigama, Thapelo Samuel Matlala

Abstract:

One of the principal challenges that the world is confronted with is insecurity. The crime rate is increasing exponentially, and protecting our physical assets especially in the motorist industry, is becoming impossible when applying our own strength. The need to develop technological solutions that detect and report theft without any human interference is inevitable. This is critical, especially for vehicle owners, to ensure theft detection and speedy identification towards recovery efforts in cases where a vehicle is missing or attempted theft is taking place. The vehicle theft detection system uses Convolutional Neural Network (CNN) to recognize the driver's face captured using an installed mobile phone device. The location identification function uses a Global Positioning System (GPS) to determine the real-time location of the vehicle. Upon identification of the location, Global System for Mobile Communications (GSM) technology is used to report or notify the vehicle owner about the whereabouts of the vehicle. The installed mobile app was implemented by making use of python as it is undoubtedly the best choice in machine learning. It allows easy access to machine learning algorithms through its widely developed library ecosystem. The graphical user interface was developed by making use of JAVA as it is better suited for mobile development. Google's online database (Firebase) was used as a means of storage for the application. The system integration test was performed using a simple percentage analysis. Sixty (60) vehicle owners participated in this study as a sample, and questionnaires were used in order to establish the acceptability of the system developed. The result indicates the efficiency of the proposed system, and consequently, the paper proposes the use of the system can effectively monitor the vehicle at any given place, even if it is driven outside its normal jurisdiction. More so, the system can be used as a database to detect, locate and report missing vehicles to different security agencies.

Keywords: CNN, location identification, tracking, GPS, GSM

Procedia PDF Downloads 132

Authors: Farahnaz Karami

Abstract:

Routing is an important, challenging task in mobile ad hoc networks due to node mobility, lack of central control, unstable links, and limited resources. An ant colony has been found to be an attractive technique for routing in Mobile Ad Hoc Networks (MANETs). However, existing swarm intelligence based routing protocols find an optimal path by considering only one or two route selection metrics without considering correlations among such parameters making them unsuitable lonely for routing real time applications. Fuzzy logic combines multiple route selection parameters containing uncertain information or imprecise data in nature, but does not have multipath routing property naturally in order to provide load balancing. The objective of this paper is to design a routing algorithm using fuzzy logic and ant colony that can solve some of routing problems in mobile ad hoc networks, such as nodes energy consumption optimization to increase network lifetime, link failures rate reduction to increase packet delivery reliability and providing load balancing to optimize available bandwidth. In proposed algorithm, the path information will be given to fuzzy inference system by ants. Based on the available path information and considering the parameters required for quality of service (QoS), the fuzzy cost of each path is calculated and the optimal paths will be selected. NS2.35 simulation tools are used for simulation and the results are compared and evaluated with the newest QoS based algorithms in MANETs according to packet delivery ratio, end-to-end delay and routing overhead ratio criterions. The simulation results show significant improvement in the performance of these networks in terms of decreasing end-to-end delay, and routing overhead ratio, and also increasing packet delivery ratio.

Keywords: mobile ad hoc networks, routing, quality of service, ant colony, fuzzy logic

Procedia PDF Downloads 37

Authors: Malik Muhammad Arslan, Muneeb Ullah, Dai Shihan, Daniyal Haider, Xiaodong Yang

Abstract:

Chest X-rays are instrumental in the detection and monitoring of a wide array of diseases, including viral infections such as COVID-19, tuberculosis, pneumonia, lung cancer, and various cardiac and pulmonary conditions. To enhance the accuracy of diagnosis, artificial intelligence (AI) algorithms, particularly deep learning models like Convolutional Neural Networks (CNNs), are employed. However, these deep learning models demand a substantial and varied dataset to attain optimal precision. Generative Adversarial Networks (GANs) can be employed to create new data, thereby supplementing the existing dataset and enhancing the accuracy of deep learning models. Nevertheless, GANs have their limitations, such as issues related to stability, convergence, and the ability to distinguish between authentic and fabricated data. In order to overcome these challenges and advance the detection and classification of CXR normal and abnormal images, this study introduces a distinctive technique known as DCWGAN (Diverse Conditional Wasserstein GAN) for generating synthetic chest X-ray (CXR) images. The study evaluates the effectiveness of this Idiosyncratic DCWGAN technique using the ResNet50 model and compares its results with those obtained using the traditional GAN approach. The findings reveal that the ResNet50 model trained on the DCWGAN-generated dataset outperformed the model trained on the classic GAN-generated dataset. Specifically, the ResNet50 model utilizing DCWGAN synthetic images achieved impressive performance metrics with an accuracy of 0.961, precision of 0.955, recall of 0.970, and F1-Measure of 0.963. These results indicate the promising potential for the early detection of diseases in CXR images using this Inimitable approach.

Keywords: CNN, classification, deep learning, GAN, Resnet50

Procedia PDF Downloads 55

Authors: Moshmi Sangavarapu

Abstract:

The US healthcare industry has undergone tremendous digital transformation in recent years, but critical care access to lower-income ethnicities is still in its nascency. This population has historically showcased substantial hesitation to seek any medical assistance. While the lack of sufficient financial resources plays a critical role, the existing cultural and knowledge barriers also contribute significantly to widening the access gap. It is imperative to break these barriers to ensure timely access to therapeutic procedures that can save important lives! Based on ongoing research, healthcare access barriers can be best addressed by tapping the untapped potential of caregiver communities first. They play a critical role in patients’ diagnoses, building healthcare knowledge and instilling confidence in required therapeutic procedures. Recent technological advancements have opened many avenues by developing smart ways of reaching the large caregiver community. A digitized go-to-market strategy featuring connected media coupled with smart IoT devices and geo-location targeting can be collectively leveraged to reach this key audience group. AI/ML algorithms can be thoroughly trained to identify relevant data signals from users' location and browsing behavior and determine useful marketing touchpoints. The web behavior can be further assimilated with natural language processing to identify contextually relevant interest topics and decipher potential caregivers on digital avenues to serve that brand message. In conclusion, grasping the true health access journey of any lower-income ethnic group is important to design beneficial touchpoints that can alleviate patients’ concerns and allow them to break their own access barriers and opt for timely and quality healthcare.

Keywords: healthcare access, market access, diversity barriers, patient journey

Procedia PDF Downloads 27

Authors: Rindra Rantoson, Hichem Nouira, Nabil Anwer, Charyar Mehdi-Souzani

Abstract:

Multiple measurements by means of various data acquisition systems are generally required to measure the shape of freeform workpieces for accuracy, reliability and holisticity. The obtained data are aligned and fused into a common coordinate system within a registration technique involving coarse and fine registrations. Standardized iterative methods have been established for fine registration such as Iterative Closest Points (ICP) and its variants. For coarse registration, no conventional method has been adopted yet despite a significant number of techniques which have been developed in the literature to supply an automatic rough matching between data sets. Two main issues are addressed in this paper: the coarse registration and the fine registration. For coarse registration, two novel automated methods based on the exploitation of discrete curvatures are presented: an enhanced Hough Transformation (HT) and an improved Ransac Transformation. The use of curvature features in both methods aims to reduce computational cost. For fine registration, a new variant of ICP method is proposed in order to reduce registration error using curvature parameters. A specific distance considering the curvature similarity has been combined with Euclidean distance to define the distance criterion used for correspondences searching. Additionally, the objective function has been improved by combining the point-to-point (P-P) minimization and the point-to-plane (P-Pl) minimization with automatic weights. These ones are determined from the preliminary calculated curvature features at each point of the workpiece surface. The algorithms are applied on simulated and real data performed by a computer tomography (CT) system. The obtained results reveal the benefit of the proposed novel curvature-based registration methods.

Keywords: discrete curvature, RANSAC transformation, hough transformation, coarse registration, ICP variant, point-to-point and point-to-plane minimization combination, computer tomography

Procedia PDF Downloads 405

Authors: Melkamu Tadesse Getachew

Abstract:

Intelligent control systems have revolutionized biomedical engineering, advancing research and enhancing medical practice. This review paper examines the impact of intelligent control on various aspects of biomedical engineering. It analyzes how these systems enhance precision and accuracy in biomedical instrumentation, improving diagnostics, monitoring, and treatment. Integration challenges are addressed, and potential solutions are proposed. The paper also investigates the optimization of drug delivery systems through intelligent control. It explores how intelligent systems contribute to precise dosing, targeted drug release, and personalized medicine. Challenges related to controlled drug release and patient variability are discussed, along with potential avenues for overcoming them. The comparison of algorithms used in intelligent control systems in biomedical control is also reviewed. The implications of intelligent control in computational and systems biology are explored, showcasing how these systems enable enhanced analysis and prediction of complex biological processes. Challenges such as interpretability, human-machine interaction, and machine reliability are examined, along with potential solutions. Intelligent control in biomedical engineering also plays a crucial role in risk management during surgical operations. This section demonstrates how intelligent systems improve patient safety and surgical outcomes when integrated into surgical robots, augmented reality, and preoperative planning. The challenges associated with these implementations and potential solutions are discussed in detail. In summary, this review paper comprehensively explores the widespread impact of intelligent control on biomedical engineering, showing the future of human health issues promising. It discusses application areas, challenges, and potential solutions, highlighting the transformative potential of these systems in advancing research and improving medical practice.

Keywords: Intelligent control systems, biomedical instrumentation, drug delivery systems, robotic surgical instruments, Computational monitoring and modeling

Procedia PDF Downloads 15

Authors: Laura Çami (Vorpsi), Xhon Skënderi

Abstract:

In today's rapidly evolving technological landscape, the traditional principles of contract law are facing significant challenges. The emergence of new technologies, such as electronic signatures, smart contracts, and online dispute resolution mechanisms, is transforming the way contracts are formed, interpreted, and enforced. This paper examines the implications of these technological advancements on the constitutional principles of contract law. One of the fundamental principles of contract law is freedom of contract, which ensures that parties have the autonomy to negotiate and enter into contracts as they see fit. However, the use of technology in the contracting process has the potential to disrupt this principle. For example, online platforms and marketplaces often offer standard-form contracts, which may not reflect the specific needs or interests of individual parties. This raises questions about the equality of bargaining power between parties and the extent to which parties are truly free to negotiate the terms of their contracts. Another important principle of contract law is the requirement of consideration, which requires that each party receives something of value in exchange for their promise. The use of digital assets, such as cryptocurrencies, has created new challenges in determining what constitutes valuable consideration in a contract. Due to the ambiguity in this area, disagreements about the legality and enforceability of such contracts may arise. Furthermore, the use of technology in dispute resolution mechanisms, such as online arbitration and mediation, may raise concerns about due process and access to justice. The use of algorithms and artificial intelligence to determine the outcome of disputes may also raise questions about the impartiality and fairness of the process. Finally, it should be noted that there are many different and complex effects of technical improvements on the fundamental constitutional foundations of contract law. As technology continues to evolve, it will be important for policymakers and legal practitioners to consider the potential impacts on contract law and to ensure that the principles of fairness, equality, and access to justice are preserved in the contracting process.

Keywords: technological advancements, constitutional principles, contract law, smart contracts, online dispute resolution, freedom of contract

Procedia PDF Downloads 97

Authors: Khanyisile Mnguni, Muthukrishnavellaisamy Kumarasamy, Jeff C. Smithers

Abstract:

Umgeni Water is a water board that supplies most parts of KwaZulu Natal with bulk portable water. Currently, Umgeni Water is running its distribution system based on required reservoir levels and demands and does not consider the energy cost at different times of the day, number of pump switches, and background leakages. Including these constraints can reduce operational cost, energy usage, leakages, and increase performance. Optimising pump schedules can reduce energy usage and costs while adhering to hydraulic and operational constraints. Umgeni Water has installed an online hydraulic software, WaterNet Advisor, that allows running different operational scenarios prior to implementation in order to optimise the distribution system. This study will investigate operation scenarios using optimisation techniques and WaterNet Advisor for a local water distribution system. Based on studies reported in the literature, introducing pump scheduling optimisation can reduce energy usage by approximately 30% without any change in infrastructure. Including tariff structures in an optimisation problem can reduce pumping costs by 15%, while including leakages decreases cost by 10%, and pressure drop in the system can be up to 12 m. Genetical optimisation algorithms are widely used due to their ability to solve nonlinear, non-convex, and mixed-integer problems. Other methods such as branch and bound linear programming have also been successfully used. A suitable optimisation method will be chosen based on its efficiency. The objective of the study is to reduce energy usage, operational cost, and leakages, and the feasibility of optimal solution will be checked using the Waternet Advisor. This study will provide an overview of the optimisation of hydraulic networks and progress made to date in multi-objective optimisation for a selected sub-system operated by Umgeni Water.

Keywords: energy usage, pump scheduling, WaterNet Advisor, leakages

Procedia PDF Downloads 72

Authors: Michael Naderhirn, Marco Pavone

Abstract:

Problem statement, motivation, and aim of work: So far, the development of control algorithms was done by control engineers in a way that the controller would fit a specification by testing. When it comes to the certification of an autonomous car in highly complex scenarios, the challenge is much higher since such a controller must mathematically guarantee to implement the rules of the road while on the other side guarantee aspects like safety and real time executability. What if it becomes reality to solve this demanding problem by combining Formal Verification and System Theory? The aim of this work is to present a workflow to solve the above mentioned problem. Summary of the presented results / main outcomes: We show the usage of an English like language to transform the rules of the road into system specification for an autonomous car. The language based specifications are used to define system functions and interfaces. Based on that a formal model is developed which formally correctly models the specifications. On the other side, a mathematical model describing the systems dynamics is used to calculate the systems reachability set which is further used to determine the system input boundaries. Then a motion planning algorithm is applied inside the system boundaries to find an optimized trajectory in combination with the formal specification model while satisfying the specifications. The result is a control strategy which can be applied in real time independent of the scenario with a mathematical guarantee to satisfy a predefined specification. We demonstrate the applicability of the method in simulation driving scenarios and a potential certification. Originality, significance, and benefit: To the authors’ best knowledge, it is the first time that it is possible to show an automated workflow which combines a specification in an English like language and a mathematical model in a mathematical formal verified way to synthesizes a controller for potential real time applications like autonomous driving.

Keywords: formal system verification, reachability, real time controller, hybrid system

Procedia PDF Downloads 219

Authors: Halkiopoulos Constantinos, Gkintoni Evgenia, Koutsopoulou Ioanna, Antonopoulou Hera

Abstract:

E-sports and e-gaming have emerged in recent years since the increase in internet use have become universal and e-gamers are the new reality in our homes. The excessive involvement of young adults with e-sports has already been revealed and the adverse consequences have been reported in researches in the past few years, but the issue has not been fully studied yet. The present research is conducted in Greece and studies the psychological profile of video game players and provides information on personality traits, habits and emotional status that affect online gamers’ behaviors in order to help professionals and policy makers address the problem. Three standardized self-report questionnaires were administered to participants who were young male and female adults aged from 19-26 years old. The Profile of Mood States (POMS) scale was used to evaluate people’s perceptions of their everyday life mood; the personality features that can trace back to people’s habits and anticipated reactions were measured by Eysenck Personality Questionnaire (EPQ), and the Trait Emotional Intelligence Questionnaire (TEIQue) was used to measure which cognitive (gamers’ beliefs) and emotional parameters (gamers’ emotional abilities) mainly affected/ predicted gamers’ behaviors and leisure time activities?/ gaming behaviors. Data mining techniques were used to analyze the data, which resulted in machine learning algorithms that were included in the software package R. The research findings attempt to designate the effect of personality traits, emotional status and emotional intelligence influence and correlation with e-sports, gamers’ behaviors and help policy makers and stakeholders take action, shape social policy and prevent the adverse consequences on young adults. The need for further research, prevention and treatment strategies is also addressed.

Keywords: e-sports, e-gamers, personality traits, POMS, emotional intelligence, data mining, R

Procedia PDF Downloads 206

Authors: Alina Fedossova, Jose Jorge Sierra Molina

Abstract:

SIP problems are part of non-classical optimization. There are problems in which the number of variables is finite, and the number of constraints is infinite. These are semi-infinite programming problems. Most algorithms for semi-infinite programming problems reduce the semi-infinite problem to a finite one and solve it by classical methods of linear or nonlinear programming. Typically, any of the constraints or the objective function is nonlinear, so the problem often involves nonlinear programming. An investment portfolio is a set of instruments used to reach the specific purposes of investors. The risk of the entire portfolio may be less than the risks of individual investment of portfolio. For example, we could make an investment of M euros in N shares for a specified period. Let yi> 0, the return on money invested in stock i for each dollar since the end of the period (i = 1, ..., N). The logical goal here is to determine the amount xi to be invested in stock i, i = 1, ..., N, such that we maximize the period at the end of ytx value, where x = (x1, ..., xn) and y = (y1, ..., yn). For us the optimal portfolio means the best portfolio in the ratio "risk-return" to the investor portfolio that meets your goals and risk ways. Therefore, investment goals and risk appetite are the factors that influence the choice of appropriate portfolio of assets. The investment returns are uncertain. Thus we have a semi-infinite programming problem. We solve a semi-infinite optimization problem of portfolio selection using the outer approximations methods. This approach can be considered as a developed Eaves-Zangwill method applying the multi-start technique in all of the iterations for the search of relevant constraints' parameters. The stochastic outer approximations method, successfully applied previously for robotics problems, Chebyshev approximation problems, air pollution and others, is based on the optimal criteria of quasi-optimal functions. As a result we obtain mathematical model and the optimal investment portfolio when yields are not clear from the beginning. Finally, we apply this algorithm to a specific case of a Colombian bank.

Keywords: outer approximation methods, portfolio problem, semi-infinite programming, numerial solution

Procedia PDF Downloads 280