Articles
Permanent URI for this collection
Browse
Browsing Articles by Type "conference proceedings"
Now showing 1 - 10 of 10
Results Per Page
Sort Options
- PublicationA real-time automatic pothole detection system using convolution neural networks(2023)
;Bharat, Ricardo ;Ikotun, Abiodun M ;Ezugwu, Absalom E. ;Abualigah, Laith ;Shehab, MohammadDetecting a pothole can help prevent damage to your vehicle and potentially prevent an accident. Different techniques, including machine learning, deep learning models, sensor methods, stereo vision, the internet of things (IoT), and black-box cameras, have already been applied to address the problem. However, studies have shown that machine learning and deep learning techniques successfully detect potholes. However, because most of these successful attempts are peculiar to the location of the study, we found no study which has addressed the peculiarity of potholes in South Africa using a tailored-trained deep learning model. In this study, we propose using a convolutional neural network (CNN), a type of deep learning model, to address this growing problem on South African roads. To achieve this, a CNN model was designed from scratch and trained with image samples obtained from the context of the study. The classifier was adapted to distinguish between a binary class which identifies the presence or absence of potholes. Results showed a significant performance enhancement at a classification accuracy of 92.72%. The outcome of this study showed that this machine learning approach holds great potential for addressing the challenge of potholes and road bumps in the region and abroad.20 1 - PublicationAlgorithms for Pixelwise Shape Deformations Preserving Digital Convexity(2022)
; ;Kenmochi, Y ;Djerroumi, H ;Coeurjolly, D. ;Romon, P.Borel, JPIn this article, we propose algorithms for pixelwise deformations of digital convex sets preserving their convexity using the combinatorics on words to identify digital convex sets via their boundary words, namely Lyndon and Christoffel words. The notion of removable and insertable points are used with a geometric strategy for choosing one of those pixels for each deformation step. The worst-case time complexity of each deflation and inflation step, which is the atomic deformation, is also analysed.15 119 - PublicationCitizen Heritage and Geoheritage : A Sampling Campaign for Cosmogenic 36Cl Surface Exposure Dating of Glacial Deposits in Mt Parnassus National Park, Greece(2022)
; ;Koukis, N ;Tsalkoutis, K ;Giorgaras, MLeontaritis, A.DA sampling campaign for 36Cl dating of Quaternary moraines was conducted in June 2021 with an open call on social media for participation of citizens regardless of their background. The organization was jointly realized by geoscientists and mountain hiking professional guides. A multi-beneficial collaboration both for science and geoheritage was achieved which resulted in a very efficient sampling campaign. At the same time participants and professional guides were actively involved in scientific research thus gaining a deeper understanding of the geoheritage of Mt Parnassus National Park.42 14 - PublicationCybers Security Analysis and Measurement Tools Using Machine Learning Approach(2022)
; ;Ghazal, Taher M. ;Hasan, Mohammad Kamrul ;Al-Dmour, Nidal A. ;Al-Sit, Waleed T.Shayla IslamArtificial intelligence (AI) and machine learning (ML) have been used in transforming our environment and the way people think, behave, and make decisions during the last few decades [1]. In the last two decades everyone connected to the Internet either an enterprise or individuals has become concerned about the security of his/their computational resources. Cybersecurity is responsible for protecting hardware and software resources from cyber attacks e.g. viruses, malware, intrusion, eavesdropping. Cyber attacks either come from black hackers or cyber warfare units. Artificial intelligence (AI) and machine learning (ML) have played an important role in developing efficient cyber security tools. This paper presents Latest Cyber Security Tools Based on Machine Learning which are: Windows defender ATP, DarckTrace, Cisco Network Analytic, IBM QRader, StringSifter, Sophos intercept X, SIME, NPL, and Symantec Targeted Attack Analytic.Scopus© Citations 4 12 - PublicationDrones Tracking Adaptation Using Reinforcement Learning: Proximal Policy optimization(2023)
; ;Seghrouchni, Amal El Fallah ;Barbaresco, FredericThis paper presents a reinforcement learning approach for automatic adaptation of the process noise covariance (Q). The Q value plays a crucial role in estimating future state values within a Kalman filter tracking system. Proximal Policy Optimization (PPO), a state-of-the-art policy optimization algorithm, was employed to determine the optimal Q value that enhances tracking performance, as measured by Root Mean Square Error (RMSE). Our results demonstrate the successful learning capability of the PPO agent over time, enabling it to suggest the optimal Q value by effectively capturing the policy of appropriate rewards under varying environmental conditions. These outcomes were compared with those of a feed-forward neural network learning, the Castella innovation/ Q values mapping, and fixed Q values. The PPO algorithm yielded promising results. We employed the Stone Soup library to simulate ground truths, measurements, and the Kalman filter tracking process.13Scopus© Citations 1 - PublicationElectric Load Probability Density Estimation using Root-Transformed Local Linear Regression(2023)
;Elhouty, Begad B.; ;El-Fouly, Tarek H. M.Zahawi, BasharProbability density estimation of stochastic electric load is of importance nowadays in power system operations and urban planning due to the uncertainties in network demand that affects the operating states of power systems. This in turn requires accurate and reliable methods to estimate network loads, especially in distribution networks. This paper proposes employing the root-unroot method in combination with local linear regression for estimating electric load probability density. Using measured load data obtained for a range of commercial enterprises, the performance of the proposed model is compared with two kernel density estimation models and two traditional parametric models (Gaussian and Gamma) and is assessed using a variety of error metrics and statistical tests. Results confirm the accuracy of the nonparametric models over the parametric models with the root transform model performing the best across all error metrics and K-S goodness-of-fit test.10 - PublicationKnowledge-based Deep Learning for Modeling Chaotic Systems(2022)
;Elabid, Zakaria; Deep Learning has received increased attention due to its unbeatable success in many fields, such as computer vision, natural language processing, recommendation systems, and most recently in simulating multiphysics problems and predicting nonlinear dynamical systems. However, modeling and forecasting the dynamics of chaotic systems remains an open research problem since training deep learning models requires big data, which is not always available in many cases. Such deep learners can be trained from additional information obtained from simulated results and by enforcing the physical laws of the chaotic systems. This paper considers extreme events and their dynamics and proposes elegant models based on deep neural networks, called knowledge-based deep learning (KDL). Our proposed KDL can learn the complex patterns governing chaotic systems by jointly training on real and simulated data directly from the dynamics and their differential equations. This knowledge is transferred to model and forecast real-world chaotic events exhibiting extreme behavior. We validate the efficiency of our model by assessing it on three real-world benchmark datasets: El Niño sea surface temperature, San Juan Dengue viral infection, and Bjørnøya daily precipitation, all governed by extreme events' dynamics. Using prior knowledge of extreme events and physics-based loss functions to lead the neural network learning, we ensure physically consistent, generalizable, and accurate forecasting, even in a small data regime. Index Terms-Chaotic systems, long short-term memory, deep learning, extreme event modeling.30 3 - PublicationPrediction of Soil Loss in a Reservoir Watershed Using an Erosion Model and Modern Technological Tools: A Case Study of Marathon Lake, Attica in Greece(2020)
; ;Kapsimalis, V. ;Evelpidou, Niki ;Apostolopoulos, G. ;Xanthakis, M ;Xanthopoulos, GPanagiotis, SMarathon Lake is an artificial reservoir with great environmental, ecological, social, and economic significance because it was the main source of water for Athens, the capital of Greece, for many years. The present study details the first attempt to map sedimentation in Marathon Lake in detail, using bathymetric mapping and soil erosion field surveying of the torrent watershed areas. First, the results of a bathymetric survey carried out in 2011 were compared with topographic maps that pre-date the construction of the dam. Based on this comparison, an estimated 8.34 hm3 of sediment have been deposited in the 80 years since the dam’s construction. In the current survey, the Revised Universal Soil Loss Equation (RUSLE) was used to estimate soil loss in the watershed area of the streams that end in Marathon Lake. The estimated value from the RUSLE was substantially lower (3.02 hm3) than that calculated in the bathymetric survey.99 106 - PublicationSliding Window Neural Generated Tracking Based on Measurement Model(2023)
; ;Seghrouchni, Amal El Fallah ;Barbaresco, FredericThis paper presents the outcome of several machine learning techniques used for the task of bird/drone classification based on their tracks. Instead of using static images, the dynamics and features extracted from the trajectories captured in videos are used to provide a more accurate and reliable recognition task. Standard Machine Learning methods such as SVM and Random Forest are used for learning this classification. Features based on the kinematics, Gabor filter, and Gray Level Co-occurrence Matrix are utilized. Several comparisons and experiments based on benchmark data sets are shown.19 - PublicationW-Transformers : A Wavelet-based Transformer Framework for Univariate Time Series Forecasting(2022)
;Sasal, Lena; Deep learning utilizing transformers has recently achieved a lot of success in many vital areas such as natural language processing, computer vision, anomaly detection, and recommendation systems, among many others. Among several merits of transformers, the ability to capture long-range temporal dependencies and interactions is desirable for time series forecasting, leading to its progress in various time series applications. In this paper, we build a transformer model for non-stationary time series. The problem is challenging yet crucially important. We present a novel framework for univariate time series representation learning based on the wavelet-based transformer encoder architecture and call it W-Transformer. The proposed W-Transformers utilize a maximal overlap discrete wavelet transformation (MODWT) to the time series data and build local transformers on the decomposed datasets to vividly capture the nonstationarity and long-range nonlinear dependencies in the time series. Evaluating our framework on several publicly available benchmark time series datasets from various domains and with diverse characteristics, we demonstrate that it performs, on average, significantly better than the baseline forecasters for short-term and long-term forecasting, even for datasets that consist of only a few hundred training samples.48 101Scopus© Citations 9