Search

An established way for improving the accuracy of gridded satellite precipitation products is to “correct” them by exploiting ground-based precipitation measurements, together with machine and statistical learning algorithms. Such corrections are made in regression settings, where the ground-based measurements are the dependent variable and the satellite data are predictor variables. Comparisons of machine and statistical learning algorithms in the direction of obtaining the most useful precipitation datasets by performing such corrections are regularly conducted in the literature. Nonetheless, in most of these comparisons, a small number of machine and statistical learning algorithms are considered. Also, small geographical regions and limited time periods are examined. Thus, the results provided tend to be of local importance and to not offer more general guidance. To provide results that are generalizable, we compared eight state-of-the-art machine and statistical learning algorithms in correcting satellite precipitation data for the entire contiguous United States and for a 15-year period. We used monthly data from the PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) gridded dataset and the Global Historical Climatology Network monthly database, version 2 (GHCNm). Our results suggest that extreme gradient boosting (XGBoost) and random forests are more accurate than the remaining algorithms, which can be ordered as follows from the best to the worst ones: Bayesian regularized feed-forward neural networks, multivariate adaptive polynomial splines (poly-MARS), gradient boosting machines (gbm), multivariate adaptive regression splines (MARS), feed-forward neural networks, linear regression.

Precise modeling of the ionospheric Total Electron Content (TEC) is critical for reliable and accurate GNSS applications. TEC is the integral of the location-dependent electron density along the signal path and is a crucial parameter that is often used to describe ionospheric variability, as it is strongly affected by solar activity. TEC is highly depended on local time (temporal variability), latitude, longitude (spatial variability), solar and geomagnetic conditions. The propagation of the signals from GNSS (Global Navigation Satellite System) satellites throughout the ionosphere is strongly influenced by temporal changes and ionospheric regular or irregular variations. Here, we leverage transformer as an effective and scalable structure with self-attention mechanisms, for modeling long-range temporal dependencies for ionospheric TEC modelling based on GNSS data. The proposed transformer model is capable of learning long-range temporal dependencies. In seq2seq models, learning temporal dependencies is a demanding task, and often the model forgets the first part, once it completes processing the whole sequence input. Our model utilizes attention mechanisms and identifies complex dependencies between input sequence elements throughout the whole sequence.Our model handles imbalanced datasets. Our work demonstrates that combining the unsupervised pre-training process with downstream task fine-tuning, offers a practical solution for ionospheric TEC modelling. This is a comparative advantage against the existing state-of-the-art works which, in most cases, fail to sufficiently model intense ionospheric variability conditions.

Satellite precipitation products are not accurate in representing the actual precipitation measured by gauges. To improve their accuracy, machine learning algorithms are applied in regression settings with ground-based measurements as dependent variables and satellite precipitation data as predictor variables. Here we examine the case of light gradient-boosting machine (LightGBM) for correcting daily IMERG (Integrated Multi-satellitE Retrievals for GPM) and PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) precipitation data using daily precipitation measurements in the contiguous US. Our demonstration especially focuses on the estimation of quantiles of the conditional probability distribution of daily precipitation at given points, with emphasis on extreme values.

Understanding the dynamics of deforestation and land uses of neighboring areas is of vital importance for the design and development of appropriate forest conservation and management policies. In this article, we approach deforestation as a multilabel classification (MLC) problem in an endeavor to capture the various relevant land uses from satellite images. To this end, we propose a multilabel vision transformer model, ForestViT, which leverages the benefits of the self-attention mechanism, obviating any convolution operations involved in commonly used deep learning models utilized for deforestation detection. Experimental evaluation in open satellite imagery datasets yields promising results in the case of MLC, particularly for imbalanced classes, and indicates ForestViT's superiority compared with well-established convolutional structures (ResNET, VGG, DenseNet, and ModileNet neural networks). This superiority is more evident for minority classes.

Merging satellite products and ground-based measurements is often required for obtaining precipitation datasets that simultaneously cover large regions with high density and are more accurate than pure satellite precipitation products. Machine and statistical learning regression algorithms are regularly utilized in this endeavor. At the same time, tree-based ensemble algorithms are adopted in various fields for solving regression problems with high accuracy and low computational costs. Still, information on which tree-based ensemble algorithm to select for correcting satellite precipitation products for the contiguous United States (US) at the daily time scale is missing from the literature. In this study, we worked towards filling this methodological gap by conducting an extensive comparison between three algorithms of the category of interest, specifically between random forests, gradient boosting machines (gbm) and extreme gradient boosting (XGBoost). We used daily data from the PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) and the IMERG (Integrated Multi-satellitE Retrievals for GPM) gridded datasets. We also used earth-observed precipitation data from the Global Historical Climatology Network daily (GHCNd) database. The experiments referred to the entire contiguous US and additionally included the application of the linear regression algorithm for benchmarking purposes. The results suggest that XGBoost is the best-performing tree-based ensemble algorithm among those compared. Indeed, the mean relative improvements that it provided with respect to linear regression (for the case that the latter algorithm was run with the same predictors as XGBoost) are equal to 52.66%, 56.26% and 64.55% (for three different predictor sets), while the respective values are 37.57%, 53.99% and 54.39% for random forests, and 34.72%, 47.99% and 62.61% for gbm. Lastly, the results suggest that IMERG is more useful than PERSIANN in the context investigated.

Land administration and management systems (LAMSs) have already made progress in the field of 3D Cadastre and the visualization of complex urban properties to support property markets and provide geospatial information for the sustainable management of smart cities. However, in less developed economies, with informally developed urban areas—the so-called self-made cities—the 2D LAMSs are left behind. Usually, they are less effective and mainly incomplete since a large number of informal constructions remain unregistered. This paper presents the latest results of an innovative on-going research aiming to structure, test and propose a low-cost but reliable enough methodology to support the simultaneous and fast implementation of both 2D land parcel and 3D property unit registration of informal, multi-story and unregistered constructions. An Indoor Positioning System (IPS) built upon low-cost Bluetooth technology combined with an innovative machine learning algorithm and connected with a 3D LADM-based cadastral mapping mobile application are the two key components of the technical solution under investigation. The proposed solution is tested for the first floor of a multi-room office building. The main conclusions concern the potential, usability and reliability of the method.

Non-intrusive load monitoring (NILM) is the task of disaggregating the total power consumption into its individual sub-components. Over the years, signal processing and machine learning algorithms have been combined to achieve this. Many publications and extensive research works are performed on energy disaggregation or NILM for the state-of-the-art methods to reach the desired performance. The initial interest of the scientific community to formulate and describe mathematically the NILM problem using machine learning tools has now shifted into a more practical NILM. Currently, we are in the mature NILM period where there is an attempt for NILM to be applied in real-life application scenarios. Thus, the complexity of the algorithms, transferability, reliability, practicality, and, in general, trustworthiness are the main issues of interest. This review narrows the gap between the early immature NILM era and the mature one. In particular, the paper provides a comprehensive literature review of the NILM methods for residential appliances only. The paper analyzes, summarizes, and presents the outcomes of a large number of recently published scholarly articles. Furthermore, the paper discusses the highlights of these methods and introduces the research dilemmas that should be taken into consideration by researchers to apply NILM methods. Finally, we show the need for transferring the traditional disaggregation models into a practical and trustworthy framework.

European and International cities face crucial global geopolitical, economic, environmental, and other changes. All these intensify threats to and inequalities in citizens' health. The implementation of Blue-Green Solutions in urban and rural areas have been broadly used to tackle the above challenges. The Mobile health (mHealth) technologies contribution in people's well-being has found to be significant. In addition, several mHealth applications have been used to support patients with mental health or cardiovascular diseases with very promising results. The patients' remote monitoring can be a valuable asset in chronic diseases management for patients suffering from diabetes, hypertension or arrhythmia, depression, asthma, allergies and others. The scope of this paper is to present the specifications, the design and the development of a mobile application which collects health-related and location data of users visiting areas with Blue-Green Solutions. The mobile application has been developed to record the citizens' and patients' physical activity and vital signs using wearable devices. The proposed application can also monitor patients physical, physiological, and emotional status as well as motivate them to engage in social and self-caring activities. Additional features include the analysis of the patients' behavior to improve self-management. The "HEART by BioAsssist" application could be used as a health and other data collection tool as well as an "intelligent assistant" to monitor and promote patient's physical activity.

Energy disaggregation, namely the separation of the aggregated household energy consumption signal into its additive sub-components, bears resemblance to the signal (source) separation problem and poses several challenges, not only as an ill-posed problem, but also, due to unsteady appliance signatures, abnormal behaviour that is usually detected in appliances operation and the existence of noise in the aggregated signal. In this paper, we propose EnerGAN++, a model based on Generative Adversarial Networks (GAN) for robust energy disaggregation. We attempt to unify the autoencoder (AE) and GAN architectures into a single framework, in which the autoencoder achieves a non-linear power signal source separation. EnerGAN++ is trained adversarially using a novel discriminator, to enhance robustness to noise. The discriminator performs sequence classification, using a recurrent convolutional neural network to handle the temporal dynamics of an appliance energy consumption time series. In particular, the proposed architecture of the discriminator leverages the ability of Convolutional Neural Networks (CNN) in rapid processing and optimal feature extraction, among with the need to infer the data temporal character and time dependence. Experimental results indicate the proposed method’s superiority compared to the current state of the art.

The urban environment seems to affect the citizens' health. The implementation of Blue-Green Solutions (BGS) in urban areas have been used to promote public health and citizens well-being. The aim of this paper is to present the development of an mHealth app for monitoring patients and citizens health status in areas where BGS will be applied. The "HEART by BioAsssist" application could be used as a health and other data collection tool as well as an "intelligent assistant" to monitor and promote patient's physical activity in areas with Blue-Green Solutions.

Data Availability Statement: All data driven applications used the our world in data COVID-19 datasets, complimented by proprietary datasets share by the STAMINA consortium. Copyright © 2022 by the authors. This paper presents the components and integrated outcome of a system that aims to achieve early detection, monitoring and mitigation of pandemic outbreaks. The architecture of the platform aims at providing a number of pandemic-response-related services, on a modular basis, that allows for the easy customization of the platform to address user’s needs per case. This customization is achieved through its ability to deploy only the necessary, loosely coupled services and tools for each case, and by providing a common authentication, data storage and data exchange infrastructure. This way, the platform can provide the necessary services without the burden of additional services that are not of use in the current deployment (e.g., predictive models for pathogens that are not endemic to the deployment area). All the decisions taken for the communication and integration of the tools that compose the platform adhere to this basic principle. The tools presented here as well as their integration is part of the project STAMINA. The paper presented is based on research undertaken as part of the European Commission-funded project STAMINA (Grant Agreement 883441).

Man overboard is an emergency in which fast and efficient detection of the critical event is the key factor for the recovery of the victim. Its severity urges the utilization of intelligent video surveillance systems that monitor the ship’s perimeter in real time and trigger the relative alarms that initiate the rescue mission. In terms of deep learning analysis, since man overboard incidents occur rarely, they present a severe class imbalance problem, and thus, supervised classification methods are not suitable. To tackle this obstacle, we follow an alternative philosophy and present a novel deep learning framework that formulates man overboard identification as an anomaly detection task. The proposed system, in the absence of training data, utilizes a multi-property spatiotemporal convolutional autoencoder that is trained only on the normal situation. We explore the use of RGB video sequences to extract specific properties of the scene, such as gradient and saliency, and utilize the autoencoders to detect anomalies. To the best of our knowledge, this is the first time that man overboard detection is made in a fully unsupervised manner while jointly learning the spatiotemporal features from RGB video streams. The algorithm achieved 97.30% accuracy and a 96.01% F1-score, surpassing the other state-of-the-art approaches significantly.

Precise modeling of the ionospheric Total Electron Content (TEC) is critical for reliable and accurate GNSS applications. TEC is the integral of the location-dependent electron density along the signal path and is a crucial parameter that is often used to describe ionospheric variability, as it is strongly affected by solar activity. TEC is highly depended on local time (temporal variability), latitude, longitude (spatial variability), solar and geomagnetic conditions. The propagation of the signals from GNSS (Global Navigation Satellite System) satellites throughout the ionosphere is strongly influenced by temporal changes and ionospheric regular or irregular variations. Here, we propose a deep learning architecture for the prediction of the vertical total electron content (VTEC) of the ionosphere based on GNSS data. The data used in many deep learning tasks until recently where mostly represented in the Euclidean space. However, geodesy studies data that have an underlying structure that is non-Euclidean space. Geospatial data are large and complex, as in the case of GNSS networks data, and their non- Euclidean nature has imposed significant challenges on the existing machine learning algorithms. The task of VTEC prediction is challenging mainly due to the complex spatiotemporal dependencies and an inherent difficulty in temporal forecasting. Spatial-temporal graph neural networks (STGNNs) aim to learn hidden patterns from spatial-temporal graphs. The key idea of STGNNs is to consider spatial and temporal dependency at the same time. Spatial Dependency: Assuming a network of permanent stations of International GNSS Service (IGS), each station represents a node of the graph, and their Euclidean distance is used to formulate the set of edges of the graph. Thus, we achieve exchange between nodes and their neighbors. Temporal dependency: The graph operates in a dynamic environment. Thus, we leverage the recurrent neural networks (RNNs) to model the temporal dependency. As a result, time series of VTEC data can be predicted to future epochs. Solar and geomagnetic indices are formulated as node attributes and are also present temporal variability.Topics to be discussed in the study include the design of the graph neural network structure, the training methods exploiting steepest descent algorithms, data analysis, as well as preliminary testing results of the VTEC predictions as compared, with state-of-the-art graph architectures.

With the rapid global urbanization, several multi-dimensional complex infrastructures have emerged, introducing new challenges in the management of the vertically stratified buildings spaces. 3D indoor cadastral spaces consist a zestful research topic as their complexity and geometry alterations during time, prevents the assignment of the corresponding Rights, Restrictions and Responsibilities (RRR). In the absence of the necessary horizontal spatial data infrastructure/floor plans their determination is weak. In this paper a fit-for-purpose technical framework and a crowdsourced methodology for the implementation of 3D cadastral surveys focused on indoor cadastral spaces, is proposed and presented. As indoor data capturing tool, an open-sourced cadastral mobile application for Android devices, is selected and presented. An Indoor Positioning System based on Bluetooth technology is established while an innovative machine learning architecture is developed, in order to explore its potentials to automatically provide the position of the mobile device within an indoor environment, aiming to add more intelligence to the proposed 3D crowdsourced cadastral framework. A practical experiment for testing the examined technical solution is conducted. The produced results are assessed to be quite promising.

Choreographic modeling, that is identification of key choreographic primitives, is a significant element for Intangible Cultural Heritage (ICH) performing art modeling. Recently, deep learning architectures, such as LSTM and CNN, have been utilized for choreographic identification and modeling. However, such approaches present sensitivity to capturing errors and fail to model the dynamic characteristics of a dance, since they assume a stationarity between the input-output data. To address these limitations, in this paper, we introduce an AutoRegressive Moving Average (ARMA) filter into a conventional CNN model; this means that the classification output feeds back to the input layer, improving overall classification accuracy. In addition, an adaptive implementation algorithm is introduced, exploiting a first-order Taylor series expansion, to update network response in order to fit dance dynamic characteristics. This way, the network parameters (e.g., weights) are dynamically modified improving overall classification accuracy. Experimental results on real-life dance sequences indicate the out-performance of the proposed approach with respect to conventional deep learning mechanisms.

This paper proposes a model suitable for predicting the ionosphere delay at different locations of receiver stations using a temporal 1D convolutional neural network (CNN) model. CNN model can optimally addresses non-linearity and model complex data through the creation of powerful representations at hierarchical levels of abstraction. To be able to predict ionosphere values for each visible satellite at a given station, sequence-to-sequence (seq2seq) models are introduced. These models are commonly used for solving sequential problems. In seq2seq models, a sequential input is entered to the model and the output has also a sequential form. Adopting this structure help us to predict ionosphere values for all satellites in view at every epoch. As experimental data, we used global navigation satellite system (GNSS) observations from selected sites in central Europe, of the global international GNSS network (IGS). The data used are part of the multi GNSS experiment (MGEX) project, that provides observations from multiple navigation satellite systems. After processing with precise point positioning (PPP) technique as implemented with GAMP software, the slant total electron content data (STEC) were obtained. The proposed CNN uses as input the ionosphere pierce points (IPP) points coordinates per visible satellite. Then, based on outcomes of the ionosphere parameters, the temporal CNN is deployed to predict future TEC variations.

Energy disaggregation, or Non-Intrusive Load Monitoring (NILM), describes various processes aiming to identify the individual contribution of appliances, given the aggregate power signal. In this paper, a non-causal adaptive context-aware bidirectional deep learning model for energy disaggregation is introduced. The proposed model, CoBiLSTM, harnesses the representational power of deep recurrent Long Short-Term Memory (LSTM) neural networks, while fitting two basic properties of NILM problem which state of the art methods do not appropriately account for: non-causality and adaptivity to contextual factors (e.g., seasonality). A Bayesian-optimized framework is introduced to select the best configuration of the proposed regression model, driven by a self-training adaptive mechanism. Furthermore, the proposed model is structured in a modular way to address multi-dimensionality issues that arise when the number of appliances increases. Experimental results indicate the proposed method’s superiority compared to the current state of the art.

In this study, we propose a tensor-based learning model to efficiently detect abnormalities on digital mammograms. Due to the fact that the availability of medical data is limited and often restricted by GDPR (general data protection regulation) compliance, the need for more sophisticated and less data-hungry approaches is urgent. Accordingly, our proposed artificial intelligence framework utilizes the canonical polyadic decomposition to decrease the trainable parameters of the wrapped Rank-R FNN model, leading to efficient learning using small amounts of data. Our model was evaluated on the open source digital mammographic database INBreast and compared with state-of-the-art models in this domain. The experimental results show that the proposed solution performs well in comparison with the other deep learning models, such as AlexNet and SqueezeNet, achieving 90% ± 4% accuracy and an F1 score of 84% ± 5%. Additionally, our framework tends to attain more robust performance with small numbers of data and is computationally lighter for inference purposes, due to the small number of trainable parameters., peer-reviewed

In Cultural Heritage, hyperspectral images are commonly used since they provide extended information regarding the optical properties of materials. Thus, the processing of such high-dimensional data becomes challenging from the perspective of machine learning techniques to be applied. In this paper, we propose a Rank-$R$ tensor-based learning model to identify and classify material defects on Cultural Heritage monuments. In contrast to conventional deep learning approaches, the proposed high order tensor-based learning demonstrates greater accuracy and robustness against overfitting. Experimental results on real-world data from UNESCO protected areas indicate the superiority of the proposed scheme compared to conventional deep learning models., Comment: Accepted for presentation in IEEE International Conference on Image Processing (ICIP 2022)

The COVID-19 pandemic has affected many aspects of human life around the world, due to its tremendous outcomes on public health and socio-economic activities. Policy makers have tried to develop efficient responses based on technologies and advanced pandemic control methodologies, to limit the wide spreading of the virus in urban areas. However, techniques such as social isolation and lockdown are short-term solutions that minimize the spread of the pandemic in cities and do not invert long-term issues that derive from climate change, air pollution and urban planning challenges that enhance the spreading ability. Thus, it seems crucial to understand what kind of factors assist or prevent the wide spreading of the virus. Although AI frameworks have a very efficient predictive ability as data-driven procedures, they often struggle to identify strong correlations among multidimensional data and provide robust explanations. In this paper, we propose the fusion of a heterogeneous, spatio-temporal dataset that combine data from eight European cities spanning from 1 January 2020 to 31 December 2021 and describe atmospheric, socio-economic, health, mobility and environmental factors all related to potential links with COVID-19. Remote sensing data are the key solution to monitor the availability on public green spaces between cities in the study period. So, we evaluate the benefits of NIR and RED bands of satellite images to calculate the NDVI and locate the percentage in vegetation cover on each city for each week of our 2-year study. This novel dataset is evaluated by a tree-based machine learning algorithm that utilizes ensemble learning and is trained to make robust predictions on daily cases and deaths. Comparisons with other machine learning techniques justify its robustness on the regression metrics RMSE and MAE. Furthermore, the explainable frameworks SHAP and LIME are utilized to locate potential positive or negative influence of the factors on global and local level, with respect to our model’s predictive ability. A variation of SHAP, namely treeSHAP, is utilized for our tree-based algorithm to make fast and accurate explanations.

Non-Intrusive Load Monitoring (NILM) describes the process of inferring the consumption pattern of appliances by only having access to the aggregated household signal. Sequence-to-sequence deep learning models have been firmly established as state-of-the-art approaches for NILM, in an attempt to identify the pattern of the appliance power consumption signal into the aggregated power signal. Exceeding the limitations of recurrent models that have been widely used in sequential modeling, this paper proposes a transformer-based architecture for NILM. Our approach, called ELECTRIcity, utilizes transformer layers to accurately estimate the power signal of domestic appliances by relying entirely on attention mechanisms to extract global dependencies between the aggregate and the domestic appliance signals. Another additive value of the proposed model is that ELECTRIcity works with minimal dataset pre-processing and without requiring data balancing. Furthermore, ELECTRIcity introduces an efficient training routine compared to other traditional transformer-based architectures. According to this routine, ELECTRIcity splits model training into unsupervised pre-training and downstream task fine-tuning, which yields performance increases in both predictive accuracy and training time decrease. Experimental results indicate ELECTRIcity’s superiority compared to several state-of-the-art methods.

Gridded satellite precipitation datasets are useful in hydrological applications as they cover large regions with high density. However, they are not accurate in the sense that they do not agree with ground-based measurements. An established means for improving their accuracy is to correct them by adopting machine learning algorithms. This correction takes the form of a regression problem, in which the ground-based measurements have the role of the dependent variable and the satellite data are the predictor variables, together with topography factors (e.g., elevation). Most studies of this kind involve a limited number of machine learning algorithms and are conducted for a small region and for a limited time period. Thus, the results obtained through them are of local importance and do not provide more general guidance and best practices. To provide results that are generalizable and to contribute to the delivery of best practices, we here compare eight state-of-the-art machine learning algorithms in correcting satellite precipitation data for the entire contiguous United States and for a 15-year period. We use monthly data from the PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) gridded dataset, together with monthly earth-observed precipitation data from the Global Historical Climatology Network monthly database, version 2 (GHCNm). The results suggest that extreme gradient boosting (XGBoost) and random forests are the most accurate in terms of the squared error scoring function. The remaining algorithms can be ordered as follows, from the best to the worst: Bayesian regularized feed-forward neural networks, multivariate adaptive polynomial splines (poly-MARS), gradient boosting machines (gbm), multivariate adaptive regression splines (MARS), feed-forward neural networks and linear regression.

Man overboard incidents in a maritime vessel are serious accidents where the rapid detection of the even is crucial for the safe retrieval of the person. To this end, the use of deep learning models as automatic detectors of these scenarios has been tested and proven efficient, however, the use of correct capturing methods is imperative in order for the learning framework to operate well. Thermal data can be a suitable method of monitoring, as they are not affected by illumination changes and are able to operate in rough conditions, such as open sea travel. We investigate the use of a convolutional autoencoder trained over thermal data, as a mechanism for the automatic detection of man overboard scenarios. Morever, we present a dataset that was created to emulate such events and was used for training and testing the algorithm.

This paper analyzes the architecture of an application programming interface (API) developed for a novel customer experience tool. The CX tool aims to monitor the customer satisfaction, based on several experience attributes and metrics, such as the Net Promoter Score. The API aims to create an efficient and user-friendly environment, which allow users to utilize all the available features of the customer experience system, including the exploitation of state-of-the-art machine learning algorithms, the analysis of the data and the graphical representation of the results.

In this paper, we scrutinize the effectiveness of various clustering techniques, investigating their applicability in Cultural Heritage monitoring applications. In the context of this paper, we detect the level of decomposition and corrosion on the walls of Saint Nicholas fort in Rhodes utilizing hyperspectral images. A total of 6 different clustering approaches have been evaluated over a set of 14 different orthorectified hyperspectral images. Experimental setup in this study involves K-means, Spectral, Meanshift, DBSCAN, Birch and Optics algorithms. For each of these techniques we evaluate its performance by the use of performance metrics such as Calinski-Harabasz, Davies-Bouldin indexes and Silhouette value. In this approach, we evaluate the outcomes of the clustering methods by comparing them with a set of annotated images which denotes the ground truth regarding the decomposition and/or corrosion area of the original images. The results depict that a few clustering techniques applied on the given dataset succeeded decent accuracy, precision, recall and f1 scores. Eventually, it was observed that the deterioration was detected quite accurately.

Ionospheric variability and disturbances can affect technologies in space and on Earth, disrupting satellite operations, communications networks, and navigation systems. The availability of numerous satellites deployed by GPS, GLONASS, Galileo, BeiDou navigation systems allows continuous monitoring of the Earth's ionosphere using measurements from these satellites. Here, we scrutinize the effectiveness and efficiency of a convolutional enriched recurrent neural network for spatio-temporal VTEC prediction. In our analysis, we have chosen different years under different solar and geomagnetic activity. We test our models for different days and at various latitudes to see model's response in cases of high ionosphere activity. Our experiments indicate that the proposed combined deep CNN-GRU model is capable of providing an accurate prediction of TEC values even in intense conditions.

In January, 2020, a new virus, named SARS-COV-2, was identified and announced to the public; in March the World Health Organization (WHO) declared a worldwide pandemic. To reduce the transmissibility of the new virus, the local authorities, worldwide, introduced a series of measures to flatten the curve. Many of the measures included some form of lockdown and movement restrictions. This unique coordinated worldwide reaction, created an opportunity for researching the effects of low traffic in air quality. In this work we research the relation between the COVID-19 measures and the Air Quality Index (AQI), using four pollutant gases (CO, O3, NO2, SO2). Also, we used a variety of machine learning tools (DNN, DTR, K-NN, Lasso, LReg, MAdam, MGBR, RFR, Ridge) to estimate the accuracy of each method in the prediction of the concentration for each gas one week later. The results showed that after the strict COVID-19 restriction measures the concentration of each pollutant gas reduced rapidly and increased again after the relaxation of lockdown measures. Finally in cases like Australia, where the measures weren’t as strict as other countries, no improvement observed.

Recent studies indicated that detecting radiographic patterns on CT chest scans can yield high sensitivity and specificity for COVID-19 detection. In this work, we scrutinize the effectiveness of deep learning models for semantic segmentation of pneumonia infected area segmentation in CT images for the detection of COVID-19. We explore the efficacy of U-Nets and Fully Convolutional Neural Networks in this task using real-world CT data from COVID-19 patients. The results indicate that Fully Convolutional Neural Networks are capable of accurate segmentation despite the class imbalance on the dataset and the man-made annotation errors on the boundaries of symptom manifestation areas, and can be a promising method for further analysis of COVID-19 induced pneumonia symptoms in CT images.Impact StatementFully Convolutional Neural Networks appear to be an accurate segmentation method in CT scans for COVID-19 pneumonia and could assist in the detection as a fast and cost-effective option.

The problem of separating the household aggregated power signal into its additive sub-components, called energy (power) disaggregation or Non-Intrusive Load Monitoring (NILM) can play an instrumental role as a driver towards consumer energy consumption awareness and behavioral change. In this paper, we propose EnerGAN++, an adversarially trained model for robust energy disaggregation. We propose a unified autoencoder (AE) and GAN architecture, in which the AE achieves a non-linear power signal source separation. The discriminator performs sequence classification, using a recurrent CNN to handle the temporal dynamics of an appliance energy consumption time series. Experimental results indicate the proposed method’s superiority compared to the state of the art.

In this paper, a crack detection mechanism for concrete tunnel surfaces is presented. The proposed methodology leverages deep Convolutional Neural Networks and domain-specific heuristic post-processing techniques to address a variety of challenges, including high accuracy requirements, low operational times and limited hardware resources, poor and variable lighting conditions, low textured lining surfaces, scarcity of training data, and abundance of noise. The proposed framework leverages the representational power of the convolutional layers of CNNs, which inherently selects effective features, thus obviating the need for the tedious task of handcrafted feature extraction. Additionally, the good performance rates attained by the proposed framework are acquired at a significantly lower execution time compared to other techniques. The presented mechanism was designed and developed as a core component of an autonomous robotic inspector deployed and validated in the tunnels of Egnatia Motorway in Metsovo, Greece. The obtained results denote the proposed approach’s superiority over a variety of methods and suggest a promising potential as a driver of autonomous concrete-lining tunnel-inspection robots.

Accurate prediction of the seawater intrusion extent is necessary for many applications, such as groundwater management or protection of coastal aquifers from water quality deterioration. However, most applications require a large number of simulations usually at the expense of prediction accuracy. In this study, the Gaussian process regression method is investigated as a potential surrogate model for the computationally expensive variable density model. Gaussian process regression is a nonparametric kernel-based probabilistic model able to handle complex relations between input and output. In this study, the extent of seawater intrusion is represented by the location of the 0.5 kg/m3 iso-chlore at the bottom of the aquifer (seawater intrusion toe). The initial position of the toe, expressed as the distance of the specific line from a number of observation points across the coastline, along with the pumping rates are the surrogate model inputs, whereas the final position of the toe constitutes the output variable set. The training sample of the surrogate model consists of 4000 variable density simulations, which differ not only in the pumping rate pattern but also in the initial concentration distribution. The Latin hypercube sampling method is used to obtain the pumping rate patterns. For comparison purposes, a number of widely used regression methods are employed, specifically regression trees and Support Vector Machine regression (linear and nonlinear). A Bayesian optimization method is applied to all the regressors, to maximize their efficiency in the prediction of seawater intrusion. The final results indicate that the Gaussian process regression method, albeit more time consuming, proved to be more efficient in terms of the mean absolute error (MAE), the root mean square error (RMSE), and the coefficient of determination (R2).

Power consumption signals of household appliances are characterized by randomly occurring events (e.g. switch-on events), making timeseries modeling a demanding process. In this paper, we propose a convolutional neural network (CNN)-based architecture with inputs and outputs formed as data sequences taking into consideration an appliance's previous states for better estimation of its current state. Furthermore, the proposed model endows CNN models with a recurrent property in order to better capture energy signal interdependencies. Using a multi-channel CNN architecture fed with additional variables related to power consumption (current, reactive, and apparent power), additionally to active power, overall performance, robustness to noise and convergence times are improved. The experimental results prove the proposed method's superiority compared to the current state of the art.

This paper proposes a new photonic-based non-invasive device for managing of Diabetic Foot Ulcers (DFUs) for people suffering from diabetes. DFUs are one of the main severe complications of diabetes, which may lead to major disabilities, such as foot amputation, or even to the death. The proposed device exploits hyperspectral (HSI) and thermal imaging to measure the status of an ulcer, in contrast to the current practice where invasive biopsies are often applied. In particular, these two photonic-based imaging techniques can estimate the biomarkers of oxyhaemoglobin (HbO2) and deoxyhaemoglobin (Hb), through which the Peripheral Oxygen Saturation (SpO2) and Tissue Oxygen Saturation (StO2) is computed. These factors are very important for the early prediction and prognosis of a DFU. The device is implemented at two editions: the in-home edition suitable for patients and the PRO (professional) edition for the medical staff. The latter is equipped with active photonic tools, such as tuneable diodes, to permit detailed diagnosis and treatment of an ulcer and its progress. The device is enriched with embedding signal processing tools for noise removal and enhancing pixel accuracy using super resolution schemes. In addition, a machine learning framework is adopted, through deep learning structures, to assist the doctors and the patients in understanding the effect of the biomarkers on DFU. The device is to be validated at large scales at three European hospitals (Charité–University Hospital in Berlin, Germany; Attikon in Athens, Greece, and Victor Babes in Timisoara, Romania) for its efficiency and performance.

Total Electron Content (TEC) is the integral of the location-dependent electron density along the signal path and is a crucial parameter that is often used to describe ionospheric variability, as it is strongly affected by solar activity. TEC is highly depended on local time, latitude, longitude, season, solar and geomagnetic conditions. The propagation of the signals from GNSS (Global Navigation Satellite System) throughout the ionosphere is strongly influenced by short- and long-term changes and ionospheric regular or irregular variations. Long short-term memory network (LSTM) is a specific recurrent neural network architecture and is capable of learning time dependence in sequential problems and can successfully model ionosphere variability. As LSTM networks “memorize” long term correlations in a sequence, they can model complex sequences with various features, where solar radio flux at 10.7 cm and magnetic activity indices are taken into consideration to provide more accurate results. Here, we propose a deep learning architecture to create regional TEC models around a station. The proposed model allows different solar and geomagnetic parameters to be inserted into the model as features. Our model has been evaluated under different solar and geomagnetic conditions. Also, the proposed model is tested for different time periods and seasonal variations and for varying geographic latitudes.

Recent studies indicate that detecting radiographic patterns on CT chest scans can yield high sensitivity and specificity for COVID-19 identification. In this paper, we scrutinize the effectiveness of deep learning models for semantic segmentation of pneumonia-infected area segmentation in CT images for the detection of COVID-19. Traditional methods for CT scan segmentation exploit a supervised learning paradigm, so they (a) require large volumes of data for their training, and (b) assume fixed (static) network weights once the training procedure has been completed. Recently, to overcome these difficulties, few-shot learning (FSL) has been introduced as a general concept of network model training using a very small amount of samples. In this paper, we explore the efficacy of few-shot learning in U-Net architectures, allowing for a dynamic fine-tuning of the network weights as new few samples are being fed into the U-Net. Experimental results indicate improvement in the segmentation accuracy of identifying COVID-19 infected regions. In particular, using 4-fold cross-validation results of the different classifiers, we observed an improvement of 5.388 ± 3.046% for all test data regarding the IoU metric and a similar increment of 5.394 ± 3.015% for the F1 score. Moreover, the statistical significance of the improvement obtained using our proposed few-shot U-Net architecture compared with the traditional U-Net model was confirmed by applying the Kruskal-Wallis test (p-value = 0.026).

In this paper, we present a multimodal deep model for detection of abnormal activity, based on bidirectional Long Short-Term Memory neural networks (LSTM). The proposed model exploits three different input modalities: RGB imagery, thermographic imagery and Channel State Information from Wi-Fi signal reflectance to estimate human intrusion and suspicious activity. The fused multimodal information is used as input in a Bidirectional LSTM, which has the benefit of being able to capture temporal interdependencies in both past and future time instances, a significant aspect in the discussed unusual activity detection scenario. We also present a Bayesian optimization framework that fine-tunes the Bidirectional LSTM parameters in an optimal manner. The proposed framework is evaluated on real-world data from a critical water infrastructure protection and monitoring scenario and the results indicate a superior performance compared to other unimodal and multimodal approaches and classification models.

An increasing number of emerging applications in data science and engineering are based on multidimensional and structurally rich data. The irregularities, however, of high-dimensional data often compromise the effectiveness of standard machine learning algorithms. We hereby propose the Rank- R Feedforward Neural Network (FNN), a tensor-based nonlinear learning model that imposes Canonical/Polyadic decomposition on its parameters, thereby offering two core advantages compared to typical machine learning methods. First, it handles inputs as multilinear arrays, bypassing the need for vectorization, and can thus fully exploit the structural information along every data dimension. Moreover, the number of the model's trainable parameters is substantially reduced, making it very efficient for small sample setting problems. We establish the universal approximation and learnability properties of Rank- R FNN, and we validate its performance on real-world hyperspectral datasets. Experimental evaluations show that Rank- R FNN is a computationally inexpensive alternative of ordinary FNN that achieves state-of-the-art performance on higher-order tensor data., peer-reviewed

In this paper, we propose a Stack Auto-encoder (SAE)-Driven and Semi-Supervised (SSL)-Based Deep Neural Network (DNN) to extract buildings from relatively low-cost satellite near infrared images. The novelty of our scheme is that we employ only an extremely small portion of labeled data for training the deep model which constitutes less than 0.08% of the total data. This way, we significantly reduce the manual effort needed to complete an annotation process, and thus the time required for creating a reliable labeled dataset. On the contrary, we apply novel semi-supervised techniques to estimate soft labels (targets) of the vast amount of existing unlabeled data and then we utilize these soft estimates to improve model training. Overall, four SSL schemes are employed, the Anchor Graph, the Safe Semi-Supervised Regression (SAFER), the Squared-loss Mutual Information Regularization (SMIR), and an equal importance Weighted Average of them (WeiAve). To retain only the most meaning information of the input data, labeled and unlabeled ones, we also employ a Stack Autoencoder (SAE) trained under an unsupervised manner. This way, we handle noise in the input signals, attributed to dimensionality redundancy, without sacrificing meaningful information. Experimental results on the benchmarked dataset of Vaihingen city in Germany indicate that our approach outperforms all state-of-the-art methods in the field using the same type of color orthoimages, though the fact that a limited dataset is utilized (10 times less data or better, compared to other approaches), while our performance is close to the one achieved by high expensive and much more precise input information like the one derived from Light Detection and Ranging (LiDAR) sensors. In addition, the proposed approach can be easily expanded to handle any number of classes, including buildings, vegetation, and ground.