Filters
total: 823
filtered: 682
Search results for: deepfm
-
Neural network agents trained by declarative programming tutors
PublicationThis paper presents an experimental study on the development of a neural network-based agent, trained using data generated using declarative programming. The focus of the study is the application of various agents to solve the classic logic task – The Wumpus World. The paper evaluates the effectiveness of neural-based agents across different map configurations, offering a comparative analysis to underline the strengths and limitations...
-
Propagation of Acoustic Disturbances in Shallow Sea
PublicationPropagation of acoustic waves in shallow sea differs fundamentally from the same phenomenon occurring in deep sea in view of non-negligible distance from the sea bottom in the first case, where presence of two regions limiting the water layer results in the acoustic pressure distribution induced by a harmonic source has an interferential nature as a result of multi-path propagation of the acoustic signal. These interferential properties...
-
Autoencoder application for anomaly detection in power consumption of lighting systems
PublicationDetecting energy consumption anomalies is a popular topic of industrial research, but there is a noticeable lack of research reported in the literature on energy consumption anomalies for road lighting systems. However, there is a need for such research because the lighting system, a key element of the Smart City concept, creates new monitoring opportunities and challenges. This paper examines algorithms based on the deep learning...
-
Vertical Temperature Stratification of the Gulf of Gdansk Water
PublicationThe Baltic Sea is characterized by variable hydroacoustic conditions, which depend on hydrological conditions throughout the year. The temperature of the water is the factor that has the greatest impact on the changes in the speed of the sound in this basin. Even at a small depth, we can observe a large temperature gradient affecting the accuracy of the conducted research using hydroacoustic devices. A characteristic feature of...
-
Robust Object Detection with Multi-input Multi-output Faster R-CNN
PublicationRecent years have seen impressive progress in visual recognition on many benchmarks, however, generalization to the out-of-distribution setting remains a significant challenge. A state-of-the-art method for robust visual recognition is model ensembling. However, recently it was shown that similarly competitive results could be achieved with a much smaller cost, by using multi-input multi-output architecture (MIMO). In this work,...
-
Age Prediction from Low Resolution, Dual-Energy X-ray Images Using Convolutional Neural Networks
PublicationAge prediction from X-rays is an interesting research topic important for clinical applications such as biological maturity assessment. It is also useful in many other practical applications, including sports or forensic investigations for age verification purposes. Research on these issues is usually carried out using high-resolution X-ray scans of parts of the body, such as images of the hands or images of the chest. In this...
-
Metal dusting phenomena of 501 AISI furnace tubes in refinery fractional distillation unit
PublicationThe purpose of this investigation was to conduct the failure analysis of 501 AISI furnace tubes places before distillation column in fractional distillation unit. The investigated furnace tubes were planned to work for ten years however after just two years of exploitation <30% of the material left. The observed corrosion process had the intense and complex character. The well-adhered shiny black deposits and deep, round pits were...
-
Speaker Recognition Using Convolutional Neural Network with Minimal Training Data for Smart Home Solutions
PublicationWith the technology advancements in smart home sector, voice control and automation are key components that can make a real difference in people's lives. The voice recognition technology market continues to involve rapidly as almost all smart home devices are providing speaker recognition capability today. However, most of them provide cloud-based solutions or use very deep Neural Networks for speaker recognition task, which are...
-
Long Distance Vital Signs Monitoring with Person Identification for Smart Home Solutions
PublicationAbstract— Imaging photoplethysmography has already been proved to be successful in short distance (below 1m). However, most of the real-life use cases of measuring vital signs require the system to work at longer distances, to be both more reliable and convenient for the user. The possible scenarios that system designers must have in mind include monitoring of the vital signs of residents in nursing homes, disabled people, who...
-
Robustness in Compressed Neural Networks for Object Detection
PublicationModel compression techniques allow to significantly reduce the computational cost associated with data processing by deep neural networks with only a minor decrease in average accuracy. Simultaneously, reducing the model size may have a large effect on noisy cases or objects belonging to less frequent classes. It is a crucial problem from the perspective of the models' safety, especially for object detection in the autonomous driving...
-
Robust Object Detection with Multi-input Multi-output Faster R-CNN
PublicationRecent years have seen impressive progress in visual recognition on many benchmarks, however, generalization to the out-of-distribution setting remains a significant challenge. A state-of-the-art method for robust visual recognition is model ensembling. However, recently it was shown that similarly competitive results could be achieved with a much smaller cost, by using multi-input multi-output architecture (MIMO). In this work,...
-
Using Long-Short term Memory networks with Genetic Algorithm to predict engine condition
PublicationPredictive maintenance (PdM) is a type of approach for maintenance processes, allowing maintenance actions to be managed depending on the machine's current condition. Maintenance is therefore carried out before failures occur. The approach doesn’t only help avoid abrupt failures but also helps lower maintenance cost and provides possibilities to manufacturers to manage maintenance budgets in a more efficient way. A new deep neural...
-
BP-EVD: Forward Block-Output Propagation for Efficient Video Denoising
PublicationDenoising videos in real-time is critical in many applications, including robotics and medicine, where varying light conditions, miniaturized sensors, and optics can substantially compromise image quality. This work proposes the first video denoising method based on a deep neural network that achieves state-of-the-art performance on dynamic scenes while running in real-time on VGA video resolution with no frame latency. The backbone...
-
Regenarative tourism – between theory and practice
PublicationPurpose: The aim of this article is to present a shift in thinking in terms of implementing the systems and practices needed to transition to a regenerative approach in tourism. The article aims to provide concrete ways to change thinking and move towards a regenerative paradigm in the tourism industry. Design/methodology/approach: This viewpoint paper defines regenerative tourism and explores its principles and the possibilities...
-
Knowledge Management and Resilience in SMEs Sector
PublicationPurpose: The aim of this paper is to investigate the role of resilience in surviving major disruptions, such as pandemic or war. This problem is especially vital for small and medium-sized enterprises (SMEs), as they often lack both resources needed for survival during prolonged economic hardship and knowledge management (KM) practices which are useful for developing the necessary business resilience. Methodology: The paper uses...
-
DNA restriction analysis as a supportive tool in mechanistic studies carried out by 32P-postlabelling
PublicationNumerous antitumor and carcinogenic compounds are able to modify DNA by forming covalent bonds with its constituents, while some anticarcinogenic compounds are known to prevent such a modification. All these processes are of vital biological import_ance, though deeper inside into factors influencing formation of DNA adducts is difficult due to the low level of their occurrence. 32P-Postlabelling approach ensures very sensitive...
-
Knowledge management strategies in KIBS companies: A preliminary analysis
PublicationPurpose – The aim of this paper is to perform a preliminary analysis concerning the detection and examination of two possible opposite approaches to KM planning which will be referred to as deliberate and emergent KM strategies. The goal is to enhance our understanding of the variety of features KM strategies possess and, accordingly, to formulate categorisations that are in line with such characteristics. Design/methodology/approach...
-
Glossary [Intellectual Output 1] Glossary as a method for reflection on complex research questions
PublicationGlobalization and digitization are strongly influencing the process of shaping the built environment. The latter is causing the new design tools to emerge faster than ever before in history, while the former is speeding up not only the development, but also the broad roll-out of more agile and interdisciplinary methodologies and work approaches. The design process is also becoming more and more inter- and trans-disciplinary. This...
-
Review on robust laser light interaction with titania – Patterning, crystallisation and ablation processes
PublicationTitanium dioxide is regarded as a very promising semiconducting material that is widely applied in many everyday-use products, devices, and processes. In general, those applications can be divided into energy or environmental categories, where a high conversion rate, and energy and power density are of particular interest. Therefore, many efforts are being put towards the elaboration of novel production routes, and improving the...
-
Architektura i woda - przekraczanie granic
PublicationPublikacja daje wgląd w procesy kształtowania połączeń wody i form zbudowanych. W obserwowanych współcześnie eksperymentach struktury budynków przeplatają się z zarysami pól wodnych, powstają konstrukcje architektoniczne unoszące się na różnego rodzaju platformach pływających. Woda jest także coraz częściej włączana w projekty urbanistyczne. Linie styku pomiędzy lądem i wodą ulegają modyfikacjom, przywraca się dawne kanały i tworzy...
-
Optymalizacja zasobów chmury obliczeniowej z wykorzystaniem inteligentnych agentów w zdalnym nauczaniu
PublicationRozprawa dotyczy optymalizacji zasobów chmury obliczeniowej, w której zastosowano inteligentne agenty w zdalnym nauczaniu. Zagadnienie jest istotne w edukacji, gdzie wykorzystuje się nowoczesne technologie, takie jak Internet Rzeczy, rozszerzoną i wirtualną rzeczywistość oraz deep learning w środowisku chmury obliczeniowej. Zagadnienie jest istotne również w sytuacji, gdy pandemia wymusza stosowanie zdalnego nauczania na dużą skalę...
-
Opracowanie technologii spawania naprawczego szyn poddźwigowych A120 oraz badania nieniszczące jakości szyn w DCT w Porcie Północnym
PublicationEkspertyza dotyczy realizacji zlecenia związanego z naprawą spoin szyn poddźwigowych na terenie terminalu kontenerowego - Deep Water Terminal (DCT) w Porcie Północnym w Gdańsku. W ramach opracowania: 1. Wykonano badania ultradźwiękowe (UT) i magnetyczno-proszkowe (MT) spoin czołowych szyn poddźwigowych typu A120, według schematu dostarczonego przez firmę SKANSKA S.A. 2. Wykonano analizę wyników badań...
-
Wpływ rodzaju zanieczyszczenia powierzchni odlewu ze staliwa LH14 na jakość napoin wykonanych elektrodą otuloną
PublicationNapawanie jest szeroko stosowaną metodą naprawiania powierzchni. Pozwala między innymi na uzupełnienie miejsc, w których uprzednio wystąpiły wżery korozyjne. Techniki napawania wykorzystywane są również do przywracania pierwotnego kształtu zużytych części. W pracy wykonano badania napoin wykonanych elektrodami otulonymi na czołowych powierzchniach cylindrycznych próbek ze staliwa LH14. Proces wykonywano dwoma rodzajami elektrod:...
-
Distortion of speech signals in the listening area: its mechanism and measurements
PublicationThe paper deals with a problem of the influence of the number and distribution of loudspeakers in speech reinforcement systems on the quality of publicly addressed voice messages, namely on speech intelligibility in the listening area. Linear superposition of time-shifted broadband waves of a same form and slightly different magnitudes that reach a listener from numerous coherent sources, is accompanied by interference effects...
-
Investigation of use of hydrophilic/hydrophobic NADESs for selective extraction of As(III) and Sb(III) ions in vegetable samples: Air assisted liquid phase microextraction and chemometric optimization
PublicationIn this paper, a green, cost-effective sample preparation method based on air assisted liquid phase microextraction (AA-LPME) was developed for the simultaneous extraction of As(III) and Sb(III) ions from vegetable samples using hydrophilic/hydrophobic natural deep eutectic solvents (NADESs). Central composite design was used for the optimization of extraction factors including NADES volume, extraction cycle, pH, and curcumin concentration....
-
Damage of a post-tensioned concrete bridge – Unwanted cracks of the girders
PublicationThe cracking of a post-tensioned T-beam superstructure, which was built using the incremental launching method, is analyzed in the paper. The problem is studied in detail, as specific damage was observed in the form of longitudinal cracks, especially in the mid-height zone of the girder at the interface of two assembly sections. The paper is a case study. A detailed inspection is done and non-destructive testing results of the...
-
Minimizing Distribution and Data Loading Overheads in Parallel Training of DNN Acoustic Models with Frequent Parameter Averaging
PublicationIn the paper we investigate the performance of parallel deep neural network training with parameter averaging for acoustic modeling in Kaldi, a popular automatic speech recognition toolkit. We describe experiments based on training a recurrent neural network with 4 layers of 800 LSTM hidden states on a 100-hour corpora of annotated Polish speech data. We propose a MPI-based modification of the training program which minimizes the...
-
Application of soil nailing technique for protection and preservation historical buildings
PublicationSoil nailing is one of the recent in situ techniques used for soil improvement and in stabilizing slopes. The process of soil nailing consists of reinforcing the natural ground with relatively small steel bars or metal rods, grouted in the pre-drilled holes. This method has a wide range of applications for stabilizing deep excavations and steep slopes. Soil nailing has recently become a very common method of slope stabilisation...
-
Medical Image Dataset Annotation Service (MIDAS)
PublicationMIDAS (Medical Image Dataset Annotation Service) is a custom-tailored tool for creating and managing datasets either for deep learning, as well as machine learning or any form of statistical research. The aim of the project is to provide one-fit-all platform for creating medical image datasets that could easily blend in hospital's workflow. In our work, we focus on the importance of medical data anonimization, discussing the...
-
Preparation and characterization of TiO2 nanostructures for catalytic CO2 photoconversion
PublicationThe titanium dioxide target (99.7%) of 1 cm in dia was ablated in vacuum by laser pulses(6 ns) at 266 nm and at repetition rate of 10 Hz. During deposition the laser fluence between 1 and 3.5 J/cm2 and the O2 pressure from the range of 10-2 - 1 Pa were applied. The thin TiO2 films were deposited on glass substrate (1 × 1 cm2) heated up to 500 °C. The chemical composition of the film and samples produced by annealing were investigated...
-
Endohedral gallide cluster superconductors and superconductivity in ReGa5
PublicationWe present transition metal-embedded (T@Gan) endohedral Ga clusters as a favorable structural motif for superconductivity and develop empirical, molecule-based, electron counting rules that govern the hierarchical architectures that the clusters assume in binary phases. Among the binary T@Gan endohedral cluster systems, Mo8Ga41, Mo6Ga31, Rh2Ga9, and Ir2 Ga9 are all previously known superconductors. The well-known exotic superconductor...
-
Bacterial community succession in an Arctic lake–stream system (Brattegg Valley, SW Spitsbergen)
PublicationThe factors affecting prokaryotic and virus structure dynamics and bacterial commu-nity composition (BCC) in aquatic habitats along a ca. 1500 m of the Brattegg Valley lake–stream system (SW Spitsbergen) composed of three postglacial lakes created by Brattegg Glacier meltwater were examined. A high number of small-volume prokaryotic cells were found in the recently-formed, deep, upper,...
-
Coherent-wave Monte Carlo method for simulating light propagation in tissue
PublicationSimulating propagation and scattering of coherent light in turbid media, such as biological tissues, is a complex problem. Numerical methods for solving Helmholtz or wave equation (e.g. finite-difference or finite-element methods) require large amount of computer memory and long computation time. This makes them impractical for simulating laser beam propagation into deep layers of tissue. Other group of methods, based on radiative...
-
The Processing Procedure for the Interpretation of Microseismic Signal Acquired from a Surface Array During Hydraulic Fracturing in Pomerania Region in Poland
PublicationHydraulic fracturing is a procedure of injecting high pressure fluid into the wellbore in order to break shell rock and facilitate gas flow. It is a very costly procedure and, if not conducted properly, it may lead to environmental pollution. To avoid costs associated with pumping fluid outside the perspective (gas rich) zone and improve one’s knowledge about the reservoir rock, microseismic monitoring can be applied. The method...
-
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...
-
Music information retrieval—The impact of technology, crowdsourcing, big data, and the cloud in art.
PublicationThe exponential growth of computer processing power, cloud data storage, and crowdsourcing model of gathering data bring new possibilities to music information retrieval (mir) field. Mir is no longer music content retrieval only; the area also comprises the discovery of expressing feelings and emotions contained in music, incorporating other than hearing modalities for helping this issue, users’ profiling, merging music with social...
-
Process Control of Biogas Purification Using Electronic Nose
PublicationNowadays, biogas produced from landfills and wastewater treatment plants or lignocellulosic biomass is important sustainable and affordable source of energy. Impurities from biogas stream can cause a serious odor problem, especially for residents of areas immediately adjacent to production plants. Therefore, biogas pre-treatment is necessary to protect engines that convert biogas into energy and in order to increase the specific...
-
Betaine and L-carnitine ester bromides: Synthesis and comparative study of their thermal behaviour and surface activity
PublicationSix esters of both betaine and L-carnitine bromides, featuring alkyl groups ranging from C8 to C18 in length, have been synthesized. The thermal behaviour of these twelve bio-based salts has been analyzed and compared by thermal gravimetric analysis and differential scanning calorimetry. The L-carnitine alkyl ester bromides melted below 100 C and can hence be considered ionic liquids (ILs) with full rights. Conversely, the betaine...
-
Text Mining Algorithms for Extracting Brand Knowledge; The fashion Industry Case
PublicationBrand knowledge is determined by customer knowledge. The opportunity to develop brands based on customer knowledge management has never been greater. Social media as a set of leading communication platforms enable peer to peer interplays between customers and brands. A large stream of such interactions is a great source of information which, when thoroughly analyzed, can become a source of innovation and lead to competitive advantage....
-
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...
-
Mussel‐inspired biomaterials: From chemistry to clinic
PublicationAfter several billions of years, nature still makes decisions on its own to identify, develop, and direct the most effective material for phenomena/challenges faced. Likewise, and inspired by the nature, we learned how to take steps in developing new technologies and materials innovations. Wet and strong adhesion by Mytilidae mussels (among which Mytilus edulis—blue mussel and Mytilus californianus—California mussel are the most...
-
Combination of homogeneous liquid–liquid extraction and vortex assisted dispersive liquid–liquid microextraction for the extraction and analysis of ochratoxin A in dried fruit samples: Central composite design optimization
PublicationThis paper presents a new analytical procedure based on combination of homogeneous liquid–liquid extraction (HLLE) and vortex-assisted dispersive liquid–liquid microextraction (VA-DLLME) for the accurate and reliable determination of ochratoxin A (OTA) in dried fruit samples. To enable selective extraction of the OTA, six hydrophobic deep eutectic solvents (hDESs) were prepared and tested as extraction solvents. Optimization of...
-
Od Safony do Sibylli : o twórczości kobiet, które miały odwagę zmieniać świat
Publication"From Sappho to Sibylla. About the women’s research and literary output preserved in the PAS Gdańsk Library early printed books collection (15th-18th c.)" The main purpose of this study was to present women’s research and literary output preserved in the early printed books collection (15th-18th c.) of the PAS Gdańsk Library (PAN Biblioteka Gdańska). The author mentions the phenomenon of women writing and participation in the...
-
Social learning in cluster initiatives
PublicationPurpose – The purpose of the paper is to portray social learning in cluster initiatives (CIs), namely: 1) to explore, with the lens of the communities of practice (CoPs) theory, in what ways social learning occurs in CIs; 2) to discover how various CoPs emerge and evolve in CIs to facilitate a collective journey in their learning process. Subsequently, the authors address the research questions: In what ways does social learning...
-
Boundary conditions for non-residential buildings from the user’s perspective: literature review
PublicationBackground and objective: This paper aims to review the boundary conditions (B/C) in specific categories (energy, building use, and lighting) within non-residential buildings to pave the way to a better understanding of users’ requirements and needs of the built environment. For this paper, B/C are understood as unique preconditions, specific characteristics for use, determining specific features of buildings, enabling an accurate...
-
Resource constrained neural network training
PublicationModern applications of neural-network-based AI solutions tend to move from datacenter backends to low-power edge devices. Environmental, computational, and power constraints are inevitable consequences of such a shift. Limiting the bit count of neural network parameters proved to be a valid technique for speeding up and increasing efficiency of the inference process. Hence, it is understandable that a similar approach is gaining...
-
CNN-CLFFA: Support Mobile Edge Computing in Transportation Cyber Physical System
PublicationIn the present scenario, the transportation Cyber Physical System (CPS) improves the reliability and efficiency of the transportation systems by enhancing the interactions between the physical and cyber systems. With the provision of better storage ability and enhanced computing, cloud computing extends transportation CPS in Mobile Edge Computing (MEC). By inspecting the existing literatures, the cloud computing cannot fulfill...
-
Investigation of antifungal and antibacterial potential of green extracts of propolis
PublicationPropolis extracts have been used in traditional medicines since ages due to its advantageous complex chemical composition. However, the antibacterial and antifungal activity of poplar propolis extracts prepared in Natural Deep Eutectic Solvent (NADES) are seldom studied. This study investigates suitable alternate for ethanol as a solvent for extraction for Polish poplar propolis. It also attempts to identify suitable extraction...
-
An LC-MS/MS Method for a Comprehensive Determination of Metabolites of BTEX Anaerobic Degradation in Bacterial Cultures and Groundwater
PublicationBTEX (benzene, toluene, ethylbenzene, and the different xylene isomers), known for carcinogenic and neurotoxic effects, are common environmental contaminants. The first step for the development of the bioremediation technologies is the detection of intense microbial degradation in contaminated waters in the quest for the most active bacterial strains. This requires the multispecies analysis for BTEX metabolites which are considered...
-
Predicting emotion from color present in images and video excerpts by machine learning
PublicationThis work aims at predicting emotion based on the colors present in images and video excerpts using a machine-learning approach. The purpose of this paper is threefold: (a) to develop a machine-learning algorithm that classifies emotions based on the color present in an image, (b) to select the best-performing algorithm from the first phase and apply it to film excerpt emotion analysis based on colors, (c) to design an online survey...