Search
Filters
total: 588
filtered: 497
Search results for: deep neural network layer
-
Efficient uncertainty quantification using sequential sampling-based neural networks
PublicationUncertainty quantification (UQ) of an engineered system involves the identification of uncertainties, modeling of the uncertainties, and the forward propagation of the uncertainties through a system analysis model. In this work, a novel surrogate-based forward propagation algorithm for UQ is proposed. The proposed algorithm is a new and unique extension of the recent efficient global optimization using neural network (NN)-based...
-
Constrained aerodynamic shape optimization using neural networks and sequential sampling
PublicationAerodynamic shape optimization (ASO) involves computational fluid dynamics (CFD)-based search for an optimal aerodynamic shape such as airfoils and wings. Gradient-based optimization (GBO) with adjoints can be used efficiently to solve ASO problems with many design variables, but problems with many constraints can still be challenging. The recently created efficient global optimization algorithm with neural network (NN)-based prediction...
-
INVESTIGATION OF THE LOMBARD EFFECT BASED ON A MACHINE LEARNING APPROACH
PublicationThe Lombard effect is an involuntary increase in the speaker’s pitch, intensity, and duration in the presence of noise. It makes it possible to communicate in noisy environments more effectively. This study aims to investigate an efficient method for detecting the Lombard effect in uttered speech. The influence of interfering noise, room type, and the gender of the person on the detection process is examined. First, acoustic parameters...
-
Limited selectivity of amperometric gas sensors operating in multicomponent gas mixtures and methods of selectivity improvement
PublicationIn recent years, smog and poor air quality have became a growing environmental problem. There is a need to continuously monitor the quality of the air. The lack of selectivity is one of the most important problems limiting the use of gas sensors for this purpose. In this study, the selectivity of six amperometric gas sensors is investigated. First, the sensors were calibrated in order to find a correlation between the concentration...
-
Vehicle Detection with Self-Training for Adaptative Video Processing Embedded Platform
PublicationTraffic monitoring from closed-circuit television (CCTV) cameras on embedded systems is the subject of the performed experiments. Solving this problem encounters difficulties related to the hardware limitations, and possible camera placement in various positions which affects the system performance. To satisfy the hardware requirements, vehicle detection is performed using a lightweight Convolutional Neural Network (CNN), named...
-
Comparison of the effectiveness of automatic EEG signal class separation algorithms
PublicationIn this paper, an algorithm for automatic brain activity class identification of EEG (electroencephalographic) signals is presented. EEG signals are gathered from seventeen subjects performing one of the three tasks: resting, watching a music video and playing a simple logic game. The methodology applied consists of several steps, namely: signal acquisition, signal processing utilizing z-score normalization, parametrization and...
-
Novel analytical method for detection of orange juice adulteration based on ultra-fast gas chromatography
PublicationThe food authenticity assessment is an increasingly important issue in food quality and safety. The application of an electronic nose based on ultra-fast gas chromatography technique enables rapid analysis of the volatile compounds from food samples. Due to the fact that this technique provides chemical profiling of natural products, it can be a powerful tool for authentication in combination with chemometrics. In this article,...
-
Determination of Odour Interactions in Gaseous Mixtures Using Electronic Nose Methods with Artificial Neural Networks
PublicationThis paper presents application of an electronic nose prototype comprised of eight sensors, five TGS-type sensors, two electrochemical sensors and one PID-type sensor, to identify odour interaction phenomenon in two-, three-, four- and five-component odorous mixtures. Typical chemical compounds, such as toluene, acetone, triethylamine, α-pinene and n-butanol, present near municipal landfills and sewage treatment plants were subjected...
-
Prediction of maximum tensile stress in plain-weave composite laminates with interacting holes via stacked machine learning algorithms: A comparative study
PublicationPlain weave composite is a long-lasting type of fabric composite that is stable enough when being handled. Open-hole composites have been widely used in industry, though they have weak structural performance and complex design processes. An extensive number of material/geometry parameters have been utilized for designing these composites, thereby an efficient computational tool is essential for that purpose. Different Machine Learning...
-
BETWEEN IDEA AND INTERPRETATION - DESIGN PROCESS AUGMENTATION
PublicationThe following paper investigates the idea of reducing the human digital intervention to a minimum during the advanced design process. Augmenting the outcome attributes beyond the designer's capabilities by computational design methods, data collection, data computing and digital fabrication, altogether imitating the human design process. The primary technical goal of the research was verification of restrictions and abilities used...
-
A taxonomy of challenges to resilient message dissemination in VANETs
PublicationInter-vehicular communications is seen as a promising solution to a number of issues related with public road safety, road congestion management, and infotainment. However, Vehicular Ad-hoc NETworks (VANETs) characterized by high mobility of vehicles and facing a number of other issues related with high frequency wireless communications and network disconnections, encounter major challenges related with reliability of message delivery....
-
Selective adsorption of BTEX on calixarene-based molecular coordination network determined by 13C NMR spectroscopy
PublicationBenzene, toluene, ethylbenzene, and xylenes (BTEX), a class of volatile organic compounds, are harmful pollutants but also very important precursors in organic industrial chemistry. Among different approaches used for the BTEX treatment, the adsorption technology has been recognized as an efficient approach because it allows to recover and reuse both adsorbent and adsorbate. However, the selective adsorption of the components is...
-
The possibility of estimating the height of the ionospheric inhomogeneities based on TEC variations maps obtained from dense GPS network
PublicationA state of the ionosphere can be effectively studied using electromagnetic signals received from global navigation satellite systems (GNSS). Utilization of the dual frequency observations allows estimating values of the total electron content (TEC). They can be used for a number of scientific studies such as detection and monitoring of traveling ionospheric disturbances or plasma bubbles. Moreover, maps of TEC variations allow...
-
The influence of small intestinal mucus structure on particle transport ex vivo
PublicationMucus provides a barrier to bacteria and toxins while allowing nutrient absorption and waste transport. Unlike colonic mucus, small intestinal mucus structure is poorly understood. This study aimed to provide evidence for a continuous, structured mucus layer and assess the diffusion of different sized particles through it. Mucus structure was assessed by histology and immunohistochemistry. Ultra-structure was assessed by scanning...
-
Badanie stanu nawierzchni drogowej z wykorzystaniem uczenia maszynowego
PublicationW artykule opisano budowę systemu informowania o stanie nawierzchni drogowej z wykorzystaniem metod cyfrowego przetwarzania obrazów oraz uczenia maszynowego. Efektem wykonanych prac badawczych jest eksperymentalna platforma, pozwalająca na rejestrację uszkodzeń na drogach, system do analizy, przetwarzania i klasyfikacji danych oraz webowa aplikacja użytkownika do przeglądu stanu nawierzchni w wybranej lokalizacji.
-
Evaluating the Use of Edge Device Towards Fall Detection in Smart City Environment
PublicationThis paper presents the development and preliminary testing of a fall detection algorithm that leverages OpenPose for real-time human pose estimation from video feeds. The system is designed to function optimally within a range of up to 7 meters from ground-level cameras, focusing exclusively on detected human silhouettes to enhance processing efficiency. The performance of the proposed approach was evaluated using accuracy values...
-
A Novel Spatio–Temporal Deep Learning Vehicle Turns Detection Scheme Using GPS-Only Data
PublicationWhether the computer is driving your car or you are, advanced driver assistance systems (ADAS) come into play on all levels, from weather monitoring to safety. These modern-day ADASs use various assisting tools for drivers to keep the journey safe; these sophisticated tools provide early signals of numerous events, such as road conditions, emerging traffic scenarios, and weather warnings. Many urban applications, such as car-sharing...
-
Applying artificial neural networks for modelling ship speed and fuel consumption
PublicationThis paper deals with modelling ship speed and fuel consumption using artificial neural network (ANN) techniques. These tools allowed us to develop ANN models that can be used for predicting both the fuel consumption and the travel time to the destination for commanded outputs (the ship driveline shaft speed and the propeller pitch) selected by the ship operator. In these cases, due to variable environmental conditions, making...
-
LONG-TERM RISK CLASS MIGRATIONS OF NON-BANKRUPT AND BANKRUPT ENTERPRISES
PublicationThis paper investigates how the process of going bankrupt can be recognized much earlier by enterprises than by traditional forecasting models. The presented studies focus on the assessment of credit risk classes and on determination of the differences in risk class migrations between non-bankrupt enterprises and future insolvent firms. For this purpose, the author has developed a model of a Kohonen artificial neural network to...
-
Sensing Direction of Human Motion Using Single-Input-Single-Output (SISO) Channel Model and Neural Networks
PublicationObject detection Through-the-Walls enables localization and identification of hidden objects behind the walls. While numerous studies have exploited Channel State Information of Multiple Input Multiple Output (MIMO) WiFi and radar devices in association with Artificial Intelligence based algorithms (AI) to detect and localize objects behind walls, this study proposes a novel non-invasive Through-the-Walls human motion direction...
-
DATABASE AND BIGDATA PROCESSING SYSTEM FOR ANALYSIS OF AIS MESSAGES IN THE NETBALTIC RESEARCH PROJECT
PublicationA specialized database and a software tool for graphical and numerical presentation of maritime measurement results has been designed and implemented as part of the research conducted under the netBaltic project (Internet over the Baltic Sea – the implementation of a multi-system, self-organizing broadband communications network over the sea for enhancing navigation safety through the development of e-navigation services.) The...
-
Modifications of silicate bioglass synthesis and composition for in vitro dissolution control: Static and dynamic assessment
PublicationA set of fifteen calcium-phosphate-silicate glass samples, varying in alkali, magnesium, silicon, and nitrogen content, was prepared, and their structural, thermal, and in vitro dissolution properties were analyzed. Infrared spectroscopy showed a high degree of depolymerization of the silicate network consisting mainly of Q2 and Q3 units. Thermal analysis showed that the silicon content primarily affects both the glass transition...
-
Novel MNZ-type microwave sensor for testing magnetodielectric materials
PublicationA novel microwave sensor with the mu-near-zero (MNZ) property is proposed for testing magnetodielectric material at 4.5 GHz. The sensor has a double-layer design consisting of a microstrip line and a metal strip with vias on layers 1 and 2, respectively. The proposed sensor can detect a unit change in relative permittivity and relative permeability with a difference in the operating frequency of 45 MHz and 78 MHz, respectively....
-
Nanocrystallization as a tool for controlling in vitro dissolution of borophosphate glass
PublicationThe controlled nanocrystallization of sodium-calcium-borophosphate glass (Na16.6Ca5.1B10.5Al0.8P10.5 O56.5 in at %) was conducted to investigate its influence on in vitro dissolution. Three temperatures (570 ◦C, 590 ◦C, and 610 ◦C) were selected based on thermal analysis and investigation of the morphology, structure, and in vitro dissolution of glass and glass-ceramics was conducted. The results of X-ray diffraction confirmed...
-
Guessing Intrinsic Forwarding Trustworthiness of Wireless Ad Hoc Network Nodes
PublicationA novel node misbehavior detection system called GIFTED is proposed for a multihop wireless ad hoc network (WAHN) whose nodes may selfishly refuse to forward transit packets. The system guesses the nodes’ intrinsic forwarding trustworthiness (IFT) by analyzing end-to-end path performance rather than utilizing unreliable and incentive incompatible low-layer mechanisms. It can work with occasional IFT jumps, directional antennae,...
-
Controlled grafting of vinylic monomers on polyolefins: a robust mathematical modeling approach
PublicationExperimental and mathematical modeling analyses were used for controlling melt free-radical grafting of vinylic monomers on polyolefins and, thereby, reducing the disturbance of undesired cross-linking of polyolefins. Response surface, desirability function, and artificial intelligence methodologies were blended to modeling/optimization of grafting reaction in terms of vinylic monomer content, peroxide initiator concentration,...
-
Self-organising maps in the analysis of strains of human abdominal wall to identify areas of similar mechanical behaviour.
PublicationThe study refers to the application of a type of artificial neural network called the Self-Organising Map (SOM) for the identification of areas of the human abdominal wall that behave in a similar mechanical way. The research is based on data acquired during in vivo tests using the digital image correlation technique (DIC). The mechanical behaviour of the human abdominal wall is analysed during changing intra-abdominal pressure....
-
Experimental tuning of AuAg nanoalloy plasmon resonances assisted by machine learning method
PublicationPlasmonic nanostructures based on AuAg nanoalloys were fabricated by thermal annealing of metallic films in an argon atmosphere. The nanoalloys were chosen because they can extend the wavelength range in which plasmon resonance occurs and thus allow the design of plasmonic platforms with the desired parameters. The influence of initial fabrication parameters and experimental conditions on the formation of nanostructures was investigated....
-
Flexible Knowledge–Vision–Integration Platform for Personal Protective Equipment Detection and Classification Using Hierarchical Convolutional Neural Networks and Active Leaning
PublicationThis work is part of an effort to develop of a Knowledge-Vision Integration Platform for Hazard Control (KVIP-HC) in industrial workplaces, adaptable to a wide range of industrial environments. The paper focuses on hazards resulted from the non-use of personal protective equipment (PPE). The objective is to test the capability of the platform to adapt to different industrial environments by simulating the process of randomly selecting...
-
Adaptacyjny system oświetlania dróg oraz inteligentnych miast
PublicationPrzedmiotem rozprawy jest zbadanie praktycznej możliwości wykrywania w czasie rzeczywistym anomalii w systemie oświetlenia drogowego w oparciu o analizę danych ze inteligentnych liczników energii. Zastosowanie inteligentnych liczników energii elektrycznej (Smart Meter) w systemach oświetlenia drogowego stwarza nowe możliwości w zakresie automatycznej diagnostyki takich niepożądanych zjawisk jak awarie lamp, odstępstwa od harmonogramu...
-
Uczenie maszynowe do samoorganizacji systemów rozproszonych w zastosowaniach gospodarczych
PublicationW pracy omówiono uczenie maszynowe do samoorganizacji systemów rozproszonych w zastosowaniach gospodarczych ze szczególnym uwzględnieniem sieci neuronowych do predykcji finansowych oraz szacowania ratingu przedsiębiorstw. Oprócz sieci neuronowych, istotną rolę w przygotowaniu i testowaniu informatycznych systemów finansowych może pełnić programowanie genetyczne. Z tego powodu omówiono uczenie maszynowe w aplikacjach konstruowanych...
-
Data-Driven Surrogate-Assisted Optimization of Metamaterial-Based Filtenna Using Deep Learning
PublicationIn this work, a computationally efficient method based on data driven surrogate models is pro-posed for the design optimization procedure of a Frequency Selective Surface (FSS)-based filtering antenna (Filtenna). A Filtenna acts as a as module that simultaneously pre-filters unwanted sig-nals, and enhances the desired signals at the operating frequency. However, due to a typically large number of design variables of FSS unit elements,...
-
The Impact of Long-Time Chemical Bonds in Mineral-Cement-Emulsion Mixtures on Stiffness Modulus
PublicationDeep cold in-place recycling is the most popular method of reuse of existing old and deteriorated asphalt layers of road pavements. In Poland, in most cases, the Mineral-Cement-Emulsion mixture technology is used, but there are also applications combining foamed bitumen and cement. Mineral-Cement-Emulsion mixtures contain two different binding agents – cement as well as asphalt from the asphalt emulsion. Asphalt creates asphalt...
-
Evaluating Accuracy of Respiratory Rate Estimation from Super Resolved Thermal Imagery
PublicationNon-contact estimation of Respiratory Rate (RR) has revolutionized the process of establishing the measurement by surpassing some issues related to attaching sensors to a body, e.g. epidermal stripping, skin disruption and pain. In this study, we perform further experiments with image processing-based RR estimation by using various image enhancement algorithms. Specifically, we employ Super Resolution (SR) Deep Learning (DL) network...
-
Smartphones as tools for equitable food quality assessment
PublicationBackground: The ubiquity of smartphones equipped with an array of sophisticated sensors, ample processing power, network connectivity and a convenient interface makes them a promising tool for non-invasive, portable food quality assessment. Combined with the recent developments in the areas of IoT, deep learning algorithms and cloud computing, they present an opportunity for advancing wide-spread, equitable and sustainable food...
-
Investigation of tetrabutylammonium bromide-glycerol-based deep eutectic solvents and their mixtures with water by spectroscopic techniques
PublicationDeep eutectic solvents (DES) are formed by an acceptor and a donor of hydrogen bonds. They are generally considered as a possible alternative to hazardous organic solvents in various fields. Very recently they have also appeared in analytical chemistry, used mainly for the separation of analytes before instrumental quantification. For the development of new extraction procedures, it is important, among other things, to understand...
-
Recognition of Emotions in Speech Using Convolutional Neural Networks on Different Datasets
PublicationArtificial Neural Network (ANN) models, specifically Convolutional Neural Networks (CNN), were applied to extract emotions based on spectrograms and mel-spectrograms. This study uses spectrograms and mel-spectrograms to investigate which feature extraction method better represents emotions and how big the differences in efficiency are in this context. The conducted studies demonstrated that mel-spectrograms are a better-suited...
-
Real-Time Facial Features Detection from Low Resolution Thermal Images with Deep Classification Models
PublicationDeep networks have already shown a spectacular success for object classification and detection for various applications from everyday use cases to advanced medical problems. The main advantage of the classification models over the detection models is less time and effort needed for dataset preparation, because classification networks do not require bounding box annotations, but labels at the image level only. Yet, after passing...
-
Urban scene semantic segmentation using the U-Net model
PublicationVision-based semantic segmentation of complex urban street scenes is a very important function during autonomous driving (AD), which will become an important technology in industrialized countries in the near future. Today, advanced driver assistance systems (ADAS) improve traffic safety thanks to the application of solutions that enable detecting objects, recognising road signs, segmenting the road, etc. The basis for these functionalities...
-
Diagnosis of damages in family buildings using neural networks
PublicationThe article concerns a problem of damages in family buildings, which result from traffic-induced vibrations. These vibrations arise from various causes and their size is influenced by many factors. The most important is the type of a road, type and weight of vehicles that run on the road, type and condition of the road surface, the distance from the house to the source of vibrations and many others which should be taken into account....
-
Wykorzystanie sztucznych sieci neuronowych do wykrywania i rozpoznawania tablic rejestracyjnych na zdjęciach pojazdów
PublicationW artykule przedstawiono koncepcję algorytmu wykrywania i rozpoznawania tablic rejestracyjnych (AWiRTR) na obrazach cyfrowych pojazdów. Detekcja i lokalizacja tablic rejestracyjnych oraz wyodrębnienie z obrazu tablicy rejestracyjnej poszczególnych znaków odbywa się z wykorzystaniem podstawowych technik przetwarzania obrazu (przekształcenia morfologiczne, wykrywanie krawędzi) jak i podstawowych danych statystycznych obiektów wykrytych...
-
Transport of Particles in Intestinal Mucus under Simulated Infant and Adult Physiological Conditions: Impact of Mucus Structure and Extracellular DNA
PublicationThe final boundary between digested food and the cells that take up nutrients in the small intestine is a protective layer of mucus. In this work, the microstructural organization and permeability of the intestinal mucus have been determined under conditions simulating those of infant and adult human small intestines. As a model, we used the mucus from the proximal (jejunal) small intestines of piglets and adult pigs. Confocal...
-
The hydrogen bond network structure within the hydration shell around simple osmolytes: Urea, tetramethylurea, and trimethylamine-N-oxide, investigated using both a fixed charge and a polarizable water model
PublicationDespite numerous experimental and computer simulation studies, a controversy still exists regarding the effect of osmolytes on the structure of surrounding water. There is a question, to what extent some of the contradictory results may arise from differences in potential models used to simulate the system or parameters employed to describe physical properties of the mixture and interpretation of the results. Bearing this in mind,...
-
The hydrogen bond network structure within the hydration shell around simple osmolytes: Urea, tetramethylurea, and trimethylamine-N-oxide, investigated using both a fixed charge and a polarizable water model
PublicationDespite numerous experimental and computer simulation studies, a controversy still exists regarding the effect of osmolytes on the structure of surrounding water. There is a question, to what extent some of the contradictory results may arise from differences in potential models used to simulate the system or parameters employed to describe physical properties of the mixture and interpretation of the results. Bearing this in mind,...
-
A novel architecture for e-learning knowledge assessment systems
PublicationIn this paper we propose a novel e-learning system, dedicated strictly to knowledge assessment tasks. In its functioning it utilizes web-based technologies, but its design differs radically from currently popular e-learning solutions which rely mostly on thin-client architecture. Our research proved that such architecture, while well suited for didactic content distribution systems is ill-suited for knowledge assessment products....
-
A novel architecture for e-learning knowledge assessment systems
PublicationAbstract. In this paper we propose a novel e-learning system, dedicated strictly to knowledge assessment tasks. In its functioning it utilizes web-based technologies, but its design differs radically from currently popular e-learning solutions which rely mostly on thin-client architecture. Our research proved that such architecture,while well suited for didactic content distribution systems is ill-suited for knowledge assessment...
-
A novel architecture for e-learning knowledge assessment systems
PublicationIn this paper we propose a novel e-learning system, dedicated strictly to knowledge assessment tasks. In its functioning it utilizes web-based technologies, but its design differs radically from currently popular e-learning solutions which rely mostly on thin-client architecture. Our research proved that such architecture, while well suited for didactic content distribution systems is ill-suited for knowledge assessment products....
-
Crystal structure and physical properties of AePd1-xP1+x (Ae = Ca, Sr)
PublicationWe report the discovery of two new compounds AePd1-xP1+x (Ae = Ca, Sr) crystallized in different hexagonal structures. Single crystals of AePd1-xP1+x (Ae = Ca, Sr) are obtained using the Bi-flux method. Crystallographic analysis by both powder and single crystal X-ray diffraction shows that CaPd1-xP1+x crystallizes in a non-centrosymmetric hexagonal structure with the space group P-6m2 (No.187) and lattice parameters a = b = 4.0391(9)...
-
A Highly Scalable, Modular Architecture for Computer Aided Assessment e-Learning Systems
PublicationIn this chapter, the authors propose a novel e-learning system, dedicated strictly to knowledge assessment tasks. In its functioning it utilizes web-based technologies, but its design differs radically from currently popular e-learning solutions which rely mostly on thin-client architecture. The authors' research proved that such architecture, while well suited for didactic content distribution systems is ill-suited for knowledge...
-
Loosely-Tied Distributed Architecture for Highly Scalable E-Learning System
PublicationVast majority of modern e-learning products are based on client-server architecture and utilization of web-based technologies (WBT). Such approach permits easy creation of e-learning systems that do not require a complex, operating system dependant client software. Unfortunately there are also drawbacks of such solution. Because of the majority of mechanisms are located on the server, its usage levels trend to build up quickly...