Search results for: DEEP%20NEURAL%20NETWORKS
-
Deep Learning Basics 2023/24
e-Learning CoursesA course about the basics of deep learning intended for students of Computer Science. It includes an introduction to supervised machine learning, the architecture of basic artificial neural networks and their training algorithms, as well as more advanced architectures (convolutional networks, recurrent networks, transformers) and regularization and optimization techniques.
-
Deep neural networks for data analysis
e-Learning CoursesThe aim of the course is to familiarize students with the methods of deep learning for advanced data analysis. Typical areas of application of these types of methods include: image classification, speech recognition and natural language understanding. Celem przedmiotu jest zapoznanie studentów z metodami głębokiego uczenia maszynowego na potrzeby zaawansowanej analizy danych. Do typowych obszarów zastosowań tego typu metod należą:...
-
Olgun Aydin dr
PeopleOlgun Aydin finished his PhD by publishing a thesis about Deep Neural Networks. He works as a Principal Machine Learning Engineer in Nike, and works as Assistant Professor in Gdansk University of Technology in Poland. Dr. Aydin is part of editorial board of "Journal of Artificial Intelligence and Data Science" Dr. Aydin served as Vice-Chairman of Why R? Foundation and is member of Polish Artificial Intelligence Society. Olgun is...
-
Deep Learning
PublicationDeep learning (DL) is a rising star of machine learning (ML) and artificial intelligence (AI) domains. Until 2006, many researchers had attempted to build deep neural networks (DNN), but most of them failed. In 2006, it was proven that deep neural networks are one of the most crucial inventions for the 21st century. Nowadays, DNN are being used as a key technology for many different domains: self-driven vehicles, smart cities,...
-
Neural networks and deep learning
PublicationIn this chapter we will provide the general and fundamental background related to Neural Networks and Deep Learning techniques. Specifically, we divide the fundamentals of deep learning in three parts, the first one introduces Deep Feed Forward Networks and the main training algorithms in the context of optimization. The second part covers Convolutional Neural Networks (CNN) and discusses their main advantages and shortcomings...
-
Experimental tests of reinforced concrete deep-beams
PublicationThe paper presents results of experimental research of the reinforced concrete deep beam with a spatial arrangement. Tested structural elements consist of the cantilever deep beam loaded on the height and transverse deep beam with hanging on it another one. The analysis includes crack morphology, effort of steel and load distribution. The article verified effectiveness of two different kind of reinforcement in both tested deep...
-
EXPERIMENTAL AND THEORETICAL FLOW OF THE FORCES IN DEEP BEAMS WITH CANTILEVAR
PublicationThis article presents the results of experimental research carried out on deep beams with cantilever which was loaded throughout the depth. The main deep beam was directly simply supported on the one side. On the other side the deep beam was suspended in another deep member situated at right angles. All deep beams created a spatial arrangement. The paper is focused on the analysis of the cracks morphology and flow of the internal...
-
A Closed Bipolar Electrochemical Cell for the Interrogation of BDD Single Particles: Electrochemical Advanced Oxidation
PublicationA closed bipolar electrochemical cell containing two conductive boron-doped diamond (BDD) particles of size 250 – 350 m, produced by high-pressure high-temperature (HPHT) synthesis, has been used to demonstrate the applicability of single BDD particles for electrochemical oxidative degradation of the dye, methylene blue (MB). The cell is fabricated using stereolithography 3D printing and the BDD particles are located at either...
-
Deep learning in the fog
PublicationIn the era of a ubiquitous Internet of Things and fast artificial intelligence advance, especially thanks to deep learning networks and hardware acceleration, we face rapid growth of highly decentralized and intelligent solutions that offer functionality of data processing closer to the end user. Internet of Things usually produces a huge amount of data that to be effectively analyzed, especially with neural networks, demands high...
-
Journal of Deep Space Exploration
Journals -
Deep Eutectic Solvents and Their Uses for Air Purification
PublicationChemical compounds released into the air by the activities of industrial plants and emitted from many other sources, including in households (paints, waxes, cosmetics, disinfectants, plastic (PVC) flooring), may affect the environment and human health. Thus, air purification is an important issue in the context of caring for the condition of the environment. Deep eutectic solvents (DESs) as liquids with environmentally friendly...
-
Supramolecular deep eutectic solvents and their applications
PublicationIn recent years, the growing awareness of the harmfulness of chemicals to the environment has resulted in the development of green and sustainable technologies. The compromise between economy and environmental requirements is based on the development of new efficient and green solutions. Supramolecular deep eutectic solvents (SUPRADESs), a new deep eutectic solvent (DES) subclass characterized by inclusion properties, are a fresh...
-
Discussion:Horizontal stress increase induced by deep vibratory compaction
PublicationDeep compaction control of granular material using the results of field tests. The analysis include the CPTU and DMT tests terformed before and after compaction works.
-
Magnetic deep eutectic solvents – Fundamentals and applications
PublicationMagnetic deep eutectic solvents (MDES), a relatively new subclass of conventional deep eutectic solvents (DES) containing additional paramagnetic components in their structure. MDES exhibit a strong response toward external magnetic fields, thus they can improve many industrial and analytical applications. In addition, this new group of solvents present unique physicochemical properties that can be easily tuned by selecting the...
-
Deep compaction control of sandy soils
PublicationVibroflotation, vibratory compaction, micro-blasting or heavy tamping are typical improvement methods for the cohesionless deposits of high thickness. The complex mechanism of deep soil compaction is related to void ratio decrease with grain rearrangements, lateral stress increase, prestressing effect of certain number of load cycles, water pressure dissipation, aging and other effects. Calibration chamber based interpretation...
-
Application of deep eutectic solvents in bioanalysis
PublicationThe application of deep eutectic solvents (DESs) is sharply surging as a green alternative to conventional solvents due to their unique properties in terms of simplicity of preparation, designability and low cost. A great deal of attention has been paid to the application of these green solvents in analytical chemistry in recent years, and a lot of interesting work has been reported. This review summarizes the most relevant applications...
-
Breast MRI segmentation by deep learning: key gaps and challenges
PublicationBreast MRI segmentation plays a vital role in early diagnosis and treatment planning of breast anomalies. Convolutional neural networks with deep learning have indicated promise in automating this process, but significant gaps and challenges remain to address. This PubMed-based review provides a comprehensive literature overview of the latest deep learning models used for breast segmentation. The article categorizes the literature...
-
NATURAL DEEP EUTECTIC SOLVENTS IN EXTRACTION PROCESS
PublicationDeveloping new, eco-friendly solvents which would meet technological and economic demands is perhaps the most popular aspects of Green Chemistry. Natural deep eutectic solvents (NADES) fully meet green chemistry principles. These solvents offer many advantages including biodegradability, low toxicity, sustainability, low costs and simple preparation. This paper provides an overview of knowledge regarding NADES with special emphasis...
-
Antifungal activity of propolis extracts produced with deep eutectic solvents.
Open Research DataThis dataset contains results of our investigation aiming in determination of antimicrobial potential of the propolis extracts produced with deep eutectic solvents. The activity was determined against C. albicans and C. glabarat strains. On the basis of these results MIC values can be calculated. Three samples of propolis were tested.
-
Deep Learning: A Case Study for Image Recognition Using Transfer Learning
PublicationDeep learning (DL) is a rising star of machine learning (ML) and artificial intelligence (AI) domains. Until 2006, many researchers had attempted to build deep neural networks (DNN), but most of them failed. In 2006, it was proven that deep neural networks are one of the most crucial inventions for the 21st century. Nowadays, DNN are being used as a key technology for many different domains: self-driven vehicles, smart cities,...
-
The influence of reinforcement on load carrying capacity and cracking of the reinforced concrete deep beam joint
PublicationThe paper presents the results of experimental research of the spatial reinforced concrete deep beam systems orthogonally reinforced and with additional inclined bars. Joint of the deep beams in this research was composed of the longitudinal deep beam with a cantilever suspended at the transversal deep beam. The cantilever deep beam was loaded throughout the depth and the transversal deep beam was loaded at the mid-span by longitudinal...
-
Deep learning for recommending subscription-limited documents
PublicationDocuments recommendation for a commercial, subscription-based online platform is important due to the difficulty in navigation through a large volume and diversity of content available to clients. However, this is also a challenging task due to the number of new documents added every day and decreasing relevance of older contents. To solve this problem, we propose deep neural network architecture that combines autoencoder with...
-
Hydrophobic deep eutectic solvents in microextraction techniques–A review
PublicationOver the past decade, deep eutectic solvents (DES) have been widely studied and applied in sample preparation techniques. Until recently, most of the synthesized DES were hydrophilic, which prevented their use in the extraction of aqueous samples. However, after 2015 studies on the synthesis and application of hydrophobic deep eutectic solvents (HDES) has rapidly expanded. Due to unique properties of HDES i.e. density, viscosity,...
-
Hydrogen bonding part sigma profile of deep eutectic solvents and pure components
Open Research DataThe set includes raw data of hydrogen bonding part sigma profiles of deep eutectic solvents and pure components generated by the ADF COSMO-RS software (SCM, Netherlands).
-
Explainable AI for Inspecting Adversarial Attacks on Deep Neural Networks
PublicationDeep Neural Networks (DNN) are state of the art algorithms for image classification. Although significant achievements and perspectives, deep neural networks and accompanying learning algorithms have some important challenges to tackle. However, it appears that it is relatively easy to attack and fool with well-designed input samples called adversarial examples. Adversarial perturba-tions are unnoticeable for humans. Such attacks...
-
Ionic Liquids and Deep Eutectic Mixtures: Sustainable Solvents for Extraction Processes
PublicationIn recent years, ionic liquids and deep eutectic mixtures have demonstrated great potential in extraction processes relevant to several scientific and technological activities. This review focuses on the applicability of these sustainable solvents in a variety of extraction techniques, including but not limited to liquid- and solid-phase (micro) extraction, microwave-assisted extraction, ultrasound-assisted extraction and pressurized...
-
Deep Eutectic Solvents: Properties and Applications in CO2 Separation
PublicationNowadays, many researchers are focused on finding a solution to the problem of global warming. Carbon dioxide is considered to be responsible for the “greenhouse” effect. The largest global emission of industrial CO2 comes from fossil fuel combustion, which makes power plants the perfect point source targets for immediate CO2 emission reductions. A state-of-the-art method for capturing carbon dioxide is chemical absorption using...
-
Training of Deep Learning Models Using Synthetic Datasets
PublicationIn order to solve increasingly complex problems, the complexity of Deep Neural Networks also needs to be constantly increased, and therefore training such networks requires more and more data. Unfortunately, obtaining such massive real world training data to optimize neural networks parameters is a challenging and time-consuming task. To solve this problem, we propose an easy-touse and general approach to training deep learning...
-
Optimized Deep Learning Model for Flood Detection Using Satellite Images
PublicationThe increasing amount of rain produces a number of issues in Kerala, particularly in urban regions where the drainage system is frequently unable to handle a significant amount of water in such a short duration. Meanwhile, standard flood detection results are inaccurate for complex phenomena and cannot handle enormous quantities of data. In order to overcome those drawbacks and enhance the outcomes of conventional flood detection...
-
Split-beam echosounder data from Gdansk Deep Summer 2019
Open Research DataThe acoustic data was collected in 2019, in the Gdansk Deep, in the season: Summer. Data was collected during the day and night. Three split-beam echosounders with frequencies of 38 kHz, 120 kHz and 333 kHz were used to collect the data. The data was collected while the ship was sailing. To ensure data quality, echosounders were calibrated and passive...
-
Benchmarking Deep Neural Network Training Using Multi- and Many-Core Processors
PublicationIn the paper we provide thorough benchmarking of deep neural network (DNN) training on modern multi- and many-core Intel processors in order to assess performance differences for various deep learning as well as parallel computing parameters. We present performance of DNN training for Alexnet, Googlenet, Googlenet_v2 as well as Resnet_50 for various engines used by the deep learning framework, for various batch sizes. Furthermore,...
-
Data augmentation for improving deep learning in image classification problem
PublicationThese days deep learning is the fastest-growing field in the field of Machine Learning (ML) and Deep Neural Networks (DNN). Among many of DNN structures, the Convolutional Neural Networks (CNN) are currently the main tool used for the image analysis and classification purposes. Although great achievements and perspectives, deep neural networks and accompanying learning algorithms have some relevant challenges to tackle. In this...
-
Are deep eutectic solvents useful in chromatography? A short review
PublicationA literature update has been done concerning Deep Eutectic Solvents (DES) use in chromatography applications. The literature survey was based on the period from 2010 till 2020 and manuscripts reported in the data bases Web of Science and Scopus. The use of DES as mobile phase and mobile phase additives, stationary phases and solid phase modifiers and the use of DES as reaction solvents for chromatography use, were evaluated. Emphasis...
-
Application of deep eutectic solvents (DES) in analytical chemistry
PublicationRecent years have been associated with efforts to reduce the impact on the natural environment. A greener approach has been introduced in various areas of science, including analytical chemistry. One of the basic procedures for preparing a sample for analysis is its extraction. Traditional methods involve the use of large amounts of organic compounds, often toxic, with an unfavorable impact on the environment. A representative...
-
The impact of the AC922 Architecture on Performance of Deep Neural Network Training
PublicationPractical deep learning applications require more and more computing power. New computing architectures emerge, specifically designed for the artificial intelligence applications, including the IBM Power System AC922. In this paper we confront an AC922 (8335-GTG) server equipped with 4 NVIDIA Volta V100 GPUs with selected deep neural network training applications, including four convolutional and one recurrent model. We report...
-
Classifying Emotions in Film Music - A Deep Learning Approach
PublicationThe paper presents an application for automatically classifying emotions in film music. A model of emotions is proposed, which is also associated with colors. The model created has nine emotional states, to which colors are assigned according to the color theory in film. Subjective tests are carried out to check the correctness of the assumptions behind the adopted emotion model. For that purpose, a statistical analysis of the...
-
The role of water in deep eutectic solvent-base extraction
PublicationDeep eutectic solvents (DESs) are currently being used in different sectors, such as electrochemistry, electrodeposition, organic synthesis, nanoparticle preparation, bioactive compound separation, etc. Their use in analytical chemistry has only recently begun to expand. Despite the publication of a sufficient number of DES-based analytical extraction procedures, some details, such as interaction of DESwith the sample and target...
-
Remarks on use of the term “deep eutectic solvent” in analytical chemistry
PublicationAbout 20 years ago, Abbott and co-workers researched new solvents that were based on mixtures of choline chloride with urea and carboxylic acids and that were liquid at ambient temperature. The term “deep eutectic solvent” (DES) was later adopted for similar mixtures. As DESs have a number of interesting features, they quickly attracted the attention of researchers and found application in various branches of chemical and materials...
-
Selected Technical Issues of Deep Neural Networks for Image Classification Purposes
PublicationIn recent years, deep learning and especially Deep Neural Networks (DNN) have obtained amazing performance on a variety of problems, in particular in classification or pattern recognition. Among many kinds of DNNs, the Convolutional Neural Networks (CNN) are most commonly used. However, due to their complexity, there are many problems related but not limited to optimizing network parameters, avoiding overfitting and ensuring good...
-
Outlier detection method by using deep neural networks
PublicationDetecting outliers in the data set is quite important for building effective predictive models. Consistent prediction can not be made through models created with data sets containing outliers, or robust models can not be created. In such cases, it may be possible to exclude observations that are determined to be outlier from the data set, or to assign less weight to these points of observation than to other points of observation....
-
Novel “acid tuned” deep eutectic solvents based on protonated L-proline
PublicationThe paper presents new types of deep eutectic solvents (DESs) based on L-proline protonated using three different acids (hydrochloric, sulfuric and phosphoric)and playing the role of a hydrogen bond acceptor(HBA). Glucose and xylitol were used as hydrogen bond donors (HBD). A series of deep eutectic solvents with various mole ratios were obtained for the systems L-proline: glucose and L-proline: xylitol. Density, melting point,...
-
Deep learning techniques for biometric security: A systematic review of presentation attack detection systems
PublicationBiometric technology, including finger vein, fingerprint, iris, and face recognition, is widely used to enhance security in various devices. In the past decade, significant progress has been made in improving biometric sys- tems, thanks to advancements in deep convolutional neural networks (DCNN) and computer vision (CV), along with large-scale training datasets. However, these systems have become targets of various attacks, with...
-
Deep eutectic solvents vs ionic liquids: Similarities and differences
PublicationDeep eutectic solvents (DES) were introduced as an alternative to ionic liquids (IL) to overcome the drawbacks of IL solvents. However, some authors consider them to be a subclass of ILs. In contrast, other authors emphasize that these are by their nature independent, different groups of substances. Thus, the question arises: Which solvent group should DESs belong to? Maybe a new class should be added to the existing ones. The...
-
Deep Learning-Based LOS and NLOS Identification in Wireless Body Area Networks
PublicationIn this article, the usage of deep learning (DL) in ultra-wideband (UWB) Wireless Body Area Networks (WBANs) is presented. The developed approach, using channel impulse response, allows higher efficiency in identifying the direct visibility conditions between nodes in off-body communication with comparison to the methods described in the literature. The effectiveness of the proposed deep feedforward neural network was checked on...
-
Force transfer and stress distribution in short cantilever deep beams loaded throughout the depth with a various reinforcement
PublicationDeep beams used as the main reinforced concrete structural elements which taking over the load and stiffening construction are often found in high-rise buildings. The architecture of these buildings is sometimes sophisticated and varied, arouse the admiration of the majority of recipients, and thus causing an engineering challenge to correctly design the structural system and force transfer. In such structures is important to shape...
-
Application of deep eutectic solvents in atomic absorption spectrometry
PublicationAtomic absorption spectrometry (AAS) is a widely applied technique for metal quantification due to its practicality, easy use and low cost. However, to improve the metrological characteristics of AAS, in particular the sensitivity and the detection limit, sample pretreatment is commonly used before the detection step itself. In consideration of the principles of Green Analytical Chemistry, new solvents are being introduced into...
-
Deep Learning-Based, Multiclass Approach to Cancer Classification on Liquid Biopsy Data
PublicationThe field of cancer diagnostics has been revolutionized by liquid biopsies, which offer a bridge between laboratory research and clinical settings. These tests are less invasive than traditional biopsies and more convenient than routine imaging methods. Liquid biopsies allow studying of tumor-derived markers in bodily fluids, enabling the development of more precise cancer diagnostic tests for screening, disease monitoring, and...
-
Nutrients, oxygen and suspended matter - Gdansk Deep (2001-2005)
Open Research DataThe results show short-term changes in the concentration of nutrients (nitrates, nitrites, ammonium ions, phosphates and total forms of nitrogen and phosphorus), dissolved oxygen and suspended particulate matter - SPM and its main components (organic carbon - POC, nitrogen - PON, phosphorus - TPP) in the water column of the Gdańsk Deep (Gdańsk Bay).
-
SYNTHESIZING MEDICAL TERMS – QUALITY AND NATURALNESS OF THE DEEP TEXT-TO-SPEECH ALGORITHM
PublicationThe main purpose of this study is to develop a deep text-to-speech (TTS) algorithm designated for an embedded system device. First, a critical literature review of state-of-the-art speech synthesis deep models is provided. The algorithm implementation covers both hardware and algorithmic solutions. The algorithm is designed for use with the Raspberry Pi 4 board. 80 synthesized sentences were prepared based on medical and everyday...
-
Deep neural network architecture search using network morphism
PublicationThe paper presents the results of the research on neural architecture search (NAS) algorithm. We utilized the hill climbing algorithm to search for well-performing structures of deep convolutional neural network. Moreover, we used the function preserving transformations which enabled the effective operation of the algorithm in a short period of time. The network obtained with the advantage of NAS was validated on skin lesion classification...
-
Challenges and Possibilities of Deep Eutectic Solvent-Based Membranes
PublicationDeep eutectic solvents (DES) are a category of a new class of solvents that can overcome some of the main drawbacks of typical solvents and ionic liquids (ILs). DES have been widely investigated and applied by the research community in several applications since their invention. Over the past years, the use of DES has been directed to the production of new materials and items for new products and processes. This is the case for...
-
Deep CNN based decision support system for detection and assessing the stage of diabetic retinopathy
PublicationThe diabetic retinopathy is a disease caused by long-standing diabetes. Lack of effective treatment can lead to vision impairment and even irreversible blindness. The disease can be diagnosed by examining digital color fundus photographs of retina. In this paper we propose deep learning approach to automated diabetic retinopathy screening. Deep convolutional neural networks (CNN) - the most popular kind of deep learning algorithms...
-
Deep learning based thermal image segmentation for laboratory animals tracking
PublicationAutomated systems for behaviour classification of laboratory animals are an attractive alternative to manual scoring. However, the proper animals separation and tracking, especially when they are in close contact, is the bottleneck of the behaviour analysis systems. In this paper, we propose a method for the segmentation of thermal images of laboratory rats that are in close contact during social behaviour tests. For this, we are...
-
Sorbents modified by deep eutectic solvents in microextraction techniques
PublicationIn recent years, considerable attention has been directed towards the employment of green solvents, specifically deep eutectic solvents (DES), in liquid phase microextraction techniques. However, comprehensive and organized knowledge regarding the modification of sorbent surface structures with DES remains limited. Therefore, this paper reviews the application of DES in modifying and improving the properties of sorbents for microextraction...
-
Decision making process using deep learning
PublicationEndüstri 4.0, dördüncü endüstri devrimi veya Endüstriyel Nesnelerin İnterneti (IIoT) olarak adlandırılan sanayi akımı, işletmelere, daha verimli, daha büyük bir esneklikle, daha güvenli ve daha çevre dostu bir şekilde üretim yapma imkanı sunmaktadır. Nesnelerin İnterneti ile bağlantılı yeni teknoloji ve hizmetler birçok endüstriyel uygulamada devrim niteliği taşımaktadır. Fabrikalardaki otomasyon, tahminleyici bakım (PdM – Predictive...
-
Deep eutectic solvents – Ideal solution for clean air or hidden danger?
PublicationThe industrial sector is one of the fastest-growing sources of greenhouse gases, due to its excessive energy consumption to meet the rapidly growing demand for energy-intensive products. The use of deep eutectic solvents (DESs) has been studied extensively in order to cope with these harmful gases, but their usage can be an issue in respect to ecological reasons. Do deep eutectic solvents harm the atmosphere? Yes, these solvents...
-
Deep Instance Segmentation of Laboratory Animals in Thermal Images
PublicationIn this paper we focus on the role of deep instance segmentation of laboratory rodents in thermal images. Thermal imaging is very suitable to observe the behaviour of laboratory animals, especially in low light conditions. It is an non-intrusive method allowing to monitor the activity of animals and potentially observe some physiological changes expressed in dynamic thermal patterns. The analysis of the recorded sequence of thermal...
-
Interpretable Deep Learning Model for the Detection and Reconstruction of Dysarthric Speech
PublicationWe present a novel deep learning model for the detection and reconstruction of dysarthric speech. We train the model with a multi-task learning technique to jointly solve dysarthria detection and speech reconstruction tasks. The model key feature is a low-dimensional latent space that is meant to encode the properties of dysarthric speech. It is commonly believed that neural networks are black boxes that solve problems but do not...
-
An Intelligent Approach to Short-Term Wind Power Prediction Using Deep Neural Networks
PublicationIn this paper, an intelligent approach to the Short-Term Wind Power Prediction (STWPP) problem is considered, with the use of various types of Deep Neural Networks (DNNs). The impact of the prediction time horizon length on accuracy, and the influence of temperature on prediction effectiveness have been analyzed. Three types of DNNs have been implemented and tested, including: CNN (Convolutional Neural Networks), GRU (Gated Recurrent...
-
The impact of the shape of deep drilled well screen openings on the filtration process in full saturation conditions
PublicationThe authors propose a supplementary method of modelling seepage flow around the deep drilled well screen. The study applies 3D numerical modelling (FEM) in order to provide an in-depth analysis of the seepage process. The analysis of filtration parameters (flow distribution q(x,t) and pressure distribution p) was conducted using the ZSoil.PC software system. The analysis indicates that the shape of perforation is of secondary importance...
-
Deep Learning Optimization for Edge Devices: Analysis of Training Quantization Parameters
PublicationThis paper focuses on convolution neural network quantization problem. The quantization has a distinct stage of data conversion from floating-point into integer-point numbers. In general, the process of quantization is associated with the reduction of the matrix dimension via limited precision of the numbers. However, the training and inference stages of deep learning neural network are limited by the space of the memory and a...
-
Microbiological condition of sediments and bottom water in the area of Gdańsk Deep in Gulf of Gdańsk
Open Research DataThis dataset contains the results of microbiological analysis of bottom water and bottom sediments in the area of Gdańsk Deep in Gulf of Gdańsk. The tested samples were collected at 5 sites on 15th of December 2007. 5 samples of bottom water and 10 samples of sediments were collected for microbiological testing. Each of these samples were analysed for...
-
Channel State Estimation in LTE-Based Heterogenous Networks Using Deep Learning
PublicationFollowing the continuous development of the information technology, the concept of dense urban networks has evolved as well. The powerful tools, like machine learning, break new ground in smart network and interface design. In this paper the concept of using deep learning for estimating the radio channel parameters of the LTE (Long Term Evolution) radio interface is presented. It was proved that the deep learning approach provides...
-
Evaluation of Facial Pulse Signals Using Deep Neural Net Models
PublicationThe reliable measurement of the pulse rate using remote photoplethysmography (PPG) is very important for many medical applications. In this paper we present how deep neural networks (DNNs) models can be used in the problem of PPG signal classification and pulse rate estimation. In particular, we show that the DNN-based classification results correspond to parameters describing the PPG signals (e.g. peak energy in the frequency...
-
Suspended matter, composition and fluxes, Gdansk Deep, late spring 2001
Open Research DataParticulate organic carbon (POC) and nitrogen (PON) concentrations and fluxes were measured in the Gdańsk Deep (Gulf of Gdansk) from 30.05 to 06.06.2001. The vertical profiles of POC and PON were characterised by the highest values in the euphotic layer, a gradual decrease with depth, and an increase below the halocline. The hydrophysical conditions...
-
Tool Wear Monitoring Using Improved Dragonfly Optimization Algorithm and Deep Belief Network
PublicationIn recent decades, tool wear monitoring has played a crucial role in the improvement of industrial production quality and efficiency. In the machining process, it is important to predict both tool cost and life, and to reduce the equipment downtime. The conventional methods need enormous quantities of human resources and expert skills to achieve precise tool wear information. To automatically identify the tool wear types, deep...
-
BIG DATA SIGNIFICANCE IN REMOTE MEDICAL DIAGNOSTICS BASED ON DEEP LEARNING TECHNIQUES
PublicationIn this paper we discuss the evaluation of neural networks in accordance with medical image classification and analysis. We also summarize the existing databases with images which could be used for training deep models that can be later utilized in remote home-based health care systems. In particular, we propose methods for remote video-based estimation of patient vital signs and other health-related parameters. Additionally, potential...
-
Position Estimation in Mixed Indoor-Outdoor Environment Using Signals of Opportunity and Deep Learning Approach
PublicationTo improve the user's localization estimation in indoor and outdoor environment a novel radiolocalization system using deep learning dedicated to work both in indoor and outdoor environment is proposed. It is based on the radio signatures using radio signals of opportunity from LTE an WiFi networks. The measurements of channel state estimators from LTE network and from WiFi network are taken by using the developed application....
-
Deep neural networks approach to skin lesions classification — A comparative analysis
PublicationThe paper presents the results of research on the use of Deep Neural Networks (DNN) for automatic classification of the skin lesions. The authors have focused on the most effective kind of DNNs for image processing, namely Convolutional Neural Networks (CNN). In particular, three kinds of CNN were analyzed: VGG19, Residual Networks (ResNet) and the hybrid of VGG19 CNN with the Support Vector Machine (SVM). The research was carried...
-
Analysis of 2D Feature Spaces for Deep Learning-based Speech Recognition
Publicationconvolutional neural network (CNN) which is a class of deep, feed-forward artificial neural network. We decided to analyze audio signal feature maps, namely spectrograms, linear and Mel-scale cepstrograms, and chromagrams. The choice was made upon the fact that CNN performs well in 2D data-oriented processing contexts. Feature maps were employed in the Lithuanian word recognition task. The spectral analysis led to the highest word...
-
Big Data from Sensor Network via Internet of Things to Edge Deep Learning for Smart City
PublicationData from a physical world is sampled by sensor networks, and then streams of Big Data are sent to cloud hosts to support decision making by deep learning software. In a smart city, some tasks may be assigned to smart devices of the Internet of Things for performing edge computing. Besides, a part of workload of calculations can be transferred to the cloud hosts. This paper proposes benchmarks for division tasks between an edge...
-
Deep slot effect in the squirrel-cage induction motors with scalar (V/F) control
PublicationQualitative characteristics of the electrical drive considerably depend on identification accuracy of math model parameters. In particular, it is depend on detection accuracy of stator active resistance r1 that is used in calculation of flux linkages, rotary speed in sensorless control systems. Paper provides analysis of influence of stator deep slot effect to stator active resistance value
-
Data, Information, Knowledge, Wisdom Pyramid Concept Revisited in the Context of Deep Learning
PublicationIn this paper, the data, information, knowledge, and wisdom (DIKW) pyramid is revisited in the context of deep learning applied to machine learningbased audio signal processing. A discussion on the DIKW schema is carried out, resulting in a proposal that may supplement the original concept. Parallels between DIWK pertaining to audio processing are presented based on examples of the case studies performed by the author and her collaborators....
-
Deep Learning Approaches in Histopathology
Publication -
Extractive detoxification of hydrolysates with simultaneous formation of deep eutectic solvents
PublicationThe hydrolysis of lignocellulosic biomass results in the production of so-called fermentation inhibitors, which reduce the efficiency of biohydrogen production. To increase the efficiency of hydrogen production, inhibitors should be removed from aqueous hydrolysate solutions before the fermentation process. This paper presents a new approach to the detoxification of hydrolysates with the simultaneous formation of in-situ deep eutectic...
-
Thermal Images Analysis Methods using Deep Learning Techniques for the Needs of Remote Medical Diagnostics
PublicationRemote medical diagnostic solutions have recently gained more importance due to global demographic shifts and play a key role in evaluation of health status during epidemic. Contactless estimation of vital signs with image processing techniques is especially important since it allows for obtaining health status without the use of additional sensors. Thermography enables us to reveal additional details, imperceptible in images acquired...
-
Charge-based deep level transient spectroscopy of B-doped and undoped polycrystalline diamond films
PublicationThe undoped and B-doped polycrystalline diamond thin film was synthesized by hot filament chemical vapor deposition and microwave plasma, respectively. The structural characterization was performed by scanning electron microscopy, X-ray diffraction and Raman spectroscopy. The electrical properties of synthesized diamond layer were characterized by dc-conductivity method and charge deep level transient spectroscopy. The B-doped...
-
Purification of model biogas from toluene using deep eutectic solvents
PublicationBiogas from landfills and wastewater treatment facilities typically contain a wide range of volatile organic compounds (VOCs), that can cause severe operational problems when biogas is used as fuel. Among the contaminants commonly occur aromatic compounds, i.e. benzene, ethylbenzene, toluene and xylenes (BTEX). In order to remove BTEX from biogas, different processes can be used. A promising process for VOCs removal is their absorption...
-
Categorization of emotions in dog behavior based on the deep neural network
PublicationThe aim of this article is to present a neural system based on stock architecture for recognizing emotional behavior in dogs. Our considerations are inspired by the original work of Franzoni et al. on recognizing dog emotions. An appropriate set of photographic data has been compiled taking into account five classes of emotional behavior in dogs of one breed, including joy, anger, licking, yawning, and sleeping. Focusing on a particular...
-
Deep learning-based waste detection in natural and urban environments
PublicationWaste pollution is one of the most significant environmental issues in the modern world. The importance of recycling is well known, both for economic and ecological reasons, and the industry demands high efficiency. Current studies towards automatic waste detection are hardly comparable due to the lack of benchmarks and widely accepted standards regarding the used metrics and data. Those problems are addressed in this article by...
-
Deep Learning-Based Intrusion System for Vehicular Ad Hoc Networks
PublicationThe increasing use of the Internet with vehicles has made travel more convenient. However, hackers can attack intelligent vehicles through various technical loopholes, resulting in a range of security issues. Due to these security issues, the safety protection technology of the in-vehicle system has become a focus of research. Using the advanced autoencoder network and recurrent neural network in deep learning, we investigated...
-
Automated detection of pronunciation errors in non-native English speech employing deep learning
PublicationDespite significant advances in recent years, the existing Computer-Assisted Pronunciation Training (CAPT) methods detect pronunciation errors with a relatively low accuracy (precision of 60% at 40%-80% recall). This Ph.D. work proposes novel deep learning methods for detecting pronunciation errors in non-native (L2) English speech, outperforming the state-of-the-art method in AUC metric (Area under the Curve) by 41%, i.e., from...
-
OPTICAL STRAIN MEASUREMENT OF CONCRETE VERSUS MANUAL EXTENSOMETER MEASUREMENT BASED ON THE TEST RC DEEP BEAM IN A COMPLEX STATE OF STRESS
PublicationThe purpose of this study is to compare the strain measurement techniques of concrete in R-C element subjected to the monotonic load up to the failure. In the analysis manual extensometer methods of measurements and the optical system ARAMIS for non-contact three-dimensional measurements of deformation was used. The test sample was a cantilever deep beam loaded throughout the depth which was a part of the reinforced concrete deep...
-
Deep eutectic solvents in analytical sample preconcentration Part B: Solid-phase (micro)extraction
PublicationOne of the key challenges of modern analytical chemistry is the monitoring of trace amounts of contaminants using sensitive and selective instrumental techniques. Due to the variety and complexity of some samples, it is often necessary to properly prepare a sample and to perform a preconcentration of trace amounts of analytes. In line with the principles of Green Analytical Chemistry (GAC), it is important for an analytical procedure...
-
Assessment and design of greener deep eutectic solvents – A multicriteria decision analysis
PublicationDeep eutectic solvents (DES) are often considered as green solvents because of their properties, such as negligible vapor pressure, biodegradability, low toxicity or natural origin of their components. Due to the fact that DES are cheaper than ionic liquids, they have gained many applications in a short period of time. However, claims about their greenness sometimes seem to be exaggerated. Especially, bearing in mind lots of data...
-
A Comprehensive Analysis of Deep Neural-Based Cerebral Microbleeds Detection System
PublicationMachine learning-based systems are gaining interest in the field of medicine, mostly in medical imaging and diagnosis. In this paper, we address the problem of automatic cerebral microbleeds (CMB) detection in magnetic resonance images. It is challenging due to difficulty in distinguishing a true CMB from its mimics, however, if successfully solved it would streamline the radiologists work. To deal with this complex three-dimensional...
-
Methods of deep modification of low-bearing soil for the foundation of new and spare air runways
PublicationAfter analyzing the impact of aircraft on the airport pavement (parking spaces, runways, startways), it was considered advisable to consider the problem of deep improvement or strengthening of its subsoil. This is especially true for low-bearing soil. The paper presents a quick and effective method of strengthening the subsoil intended for the construction of engineering structures used for civil...
-
Hydrophobic deep eutectic solvents as “green” extraction media for polycyclic aromatic hydrocarbons in aqueous samples
PublicationThe paper presents novel nonionic and hydrophobic deep eutectic solvents which were synthesized from natural compounds, i.e., thymol, ±camphor, decanoic and 10-undecylenic acids. Fundamental physicochemical properties of the synthesized deep eutectic solvents were determined, followed by their application as extractants in ultrasound-assisted dispersive liquid-liquid microextraction to isolate and enrich polycyclic aromatic hydrocarbons...
-
Efficient Extraction of Fermentation Inhibitors by Means of Green Hydrophobic Deep Eutectic Solvents
PublicationThe methods for hydrogen yield efficiency improvements, the gaseous stream purification in gaseous biofuels generation, and the biomass pretreatment are considered as the main trends in research devoted to gaseous biofuel production. The environmental aspect related to the liquid stream purification arises. Moreover, the management of post-fermentation broth with the application of various biorefining techniques gains importance....
-
Experimental and predicted physicochemical properties of monopropanolamine-based deep eutectic solvents
PublicationIn this work, the novel deep eutectic solvents (DESs) based on 3-amino-1-propanol (AP) as hydrogen bond donor (HBD) and tetrabutylammonium bromide (TBAB) or tetrabutylammonium chloride (TBAC) or tetraethylammonium chloride (TEAC) as hydrogen bond acceptors (HBAs) were synthesized with different molar ratios of 1: 4, 1: 6 and 1: 8 salt to AP. Fourier Transform Infrared Spectroscopy measurements were performed to provide an evidence...
-
Face with Mask Detection in Thermal Images Using Deep Neural Networks
PublicationAs the interest in facial detection grows, especially during a pandemic, solutions are sought that will be effective and bring more benefits. This is the case with the use of thermal imaging, which is resistant to environmental factors and makes it possible, for example, to determine the temperature based on the detected face, which brings new perspectives and opportunities to use such an approach for health control purposes. The...
-
Experimental and predicted physicochemical properties of monopropanolamine-based deep eutectic solvents
PublicationIn this work, the novel deep eutectic solvents (DESs) based on 3-amino-1-propanol (AP) as hydrogen bond donor (HBD) and tetrabutylammonium bromide (TBAB) or tetrabutylammonium chloride (TBAC) or tetraethylammonium chloride (TEAC) as hydrogen bond acceptors (HBAs) were synthesized with different molar ratios of 1:4, 1:6 and 1:8 salt to AP. Fourier Transform Infrared Spectroscopy measurements were performed to provide an evidence...
-
Closer look into the structures of tetrabutylammonium bromide–glycerol-based deep eutectic solvents and their mixtures with water
PublicationIn recent years, deep eutectic solvents (DES) and it’s mixture with water have become more and more attention as green solvents used in chemistry. However, there are only a few theoretical studies on the mechanisms of pure DES and DES-water complex formation. Therefore, the structural properties of tetrabutylammonium bromide–glycerol-based deep eutectic solvents and their mixtures with water have been investigated by means of Molecular...
-
Influence of Thermal Imagery Resolution on Accuracy of Deep Learning based Face Recognition
PublicationHuman-system interactions frequently require a retrieval of the key context information about the user and the environment. Image processing techniques have been widely applied in this area, providing details about recognized objects, people and actions. Considering remote diagnostics solutions, e.g. non-contact vital signs estimation and smart home monitoring systems that utilize person’s identity, security is a very important factor....
-
Pursuing the Deep-Learning-Based Classification of Exposed and Imagined Colors from EEG
PublicationEEG-based brain-computer interfaces are systems aiming to integrate disabled people into their environments. Nevertheless, their control could not be intuitive or depend on an active external stimulator to generate the responses for interacting with it. Targeting the second issue, a novel paradigm is explored in this paper, which depends on a passive stimulus by measuring the EEG responses of a subject to the primary colors (red,...
-
LOS and NLOS identification in real indoor environment using deep learning approach
PublicationVisibility conditions between antennas, i.e. Line-of-Sight (LOS) and Non-Line-of-Sight (NLOS) can be crucial in the context of indoor localization, for which detecting the NLOS condition and further correcting constant position estimation errors or allocating resources can reduce the negative influence of multipath propagation on wireless communication and positioning. In this paper a deep learning (DL) model to classify LOS/NLOS...
-
Solubility advantage of sulfanilamide and sulfacetamide in natural deep eutectic systems: experimental and theoretical investigations
PublicationObjective: The aim of this study was to explore the possibility of using natural deep eutectic solvents (NADES) as solvation media for enhancement of solubility of sulfonamides, as well as gaining some thermodynamic characteristics of the analyzed systems. Significance: Low solubility of many active pharmaceutical ingredients is a well-recognized difficulty in pharmaceutical industry, hence the need for different strategies addressing...
-
Deep convolutional neural network for predicting kidney tumour malignancy
PublicationPurpose: According to the statistics, up to 15-20% of removed solid kidney tumors turn out to be benign in postoperative histopathological examination, despite having been identified as malignant by a radiologist. The aim of the research was to limit the number of unnecessary nephrectomies of benign tumors. Methods or Background: We propose a machine-aided diagnostic system for kidney...
-
Improving Accuracy of Contactless Respiratory Rate Estimation by Enhancing Thermal Sequences with Deep Neural Networks
PublicationEstimation of vital signs using image processing techniques have already been proved to have a potential for supporting remote medical diagnostics and replacing traditional measurements that usually require special hardware and electrodes placed on a body. In this paper, we further extend studies on contactless Respiratory Rate (RR) estimation from extremely low resolution thermal imagery by enhancing acquired sequences using Deep...
-
Experimental and Theoretical Insights into the Intermolecular Interactions in Saturated Systems of Dapsone in Conventional and Deep Eutectic Solvents
PublicationSolubility is not only a crucial physicochemical property for laboratory practice but also provides valuable insight into the mechanism of saturated system organization, as a measure of the interplay between various intermolecular interactions. The importance of these data cannot be overstated, particularly when dealing with active pharmaceutical ingredients (APIs), such as dapsone. It is a commonly used anti-inflammatory and...