Search
Search results for: image quality
-
Neural Network Subgraphs Correlation with Trained Model Accuracy
PublicationNeural Architecture Search (NAS) is a computationally demanding process of finding optimal neural network architecture for a given task. Conceptually, NAS comprises applying a search strategy on a predefined search space accompanied by a performance evaluation method. The design of search space alone is expected to substantially impact NAS efficiency. We consider neural networks as graphs and find a correlation between the presence...
-
Imaging of Increasing Damage in Steel Plates Using Lamb Waves and Ultrasound Computed Tomography
PublicationThis paper concerns inspection of steel plates, with particular emphasis on the assessment of in-creasing damage. Non-destructive tests were performed on four plates, one of which was un-damaged, while the remaining three had a defects in the form of circular holes with a diameter of 2, 5, and 10 cm. Guided Lamb waves were used in the research, and the image reconstruction was performed using ultrasound computed tomography. The...
-
An Overview of the Development of a Real-Time System for Endoscopic Video Classification
PublicationThe article presents the results of improving endoscopic image classification algorithms in an effort towards applying them in a real-time diagnosis supporting system. Methods for the detection and removal of personal data are presented and discussed. The currently developed recognition algorithms have been improved in terms of accuracy and performance to make them suitable for a real-life implementation. Their test results are...
-
Pumping effect measured by PIV method in multi-layer spike electrode EHD device for air cleaning
PublicationDust particles can be harmful for human health when inhaled. Especially dangerous are submicron dust particles, which can contain traces of toxic elements and can easily penetrate into human respiratory system. Thus, new devices for air cleaning are needed. In this work the flow velocity field patterns measured by Particle Image Velocimetry (PIV) method in an electrohydrodynamic (EHD) device for air cleaning are presented. The...
-
Resolving Conflicts in Object Tracking in Video Stream Employing Key Point Matching
PublicationA novel approach to resolving ambiguous situations in object tracking in video streams is presented. The proposed method combines standard tracking technique employing Kalman filters with global feature matching method. Object detection is performed using a background subtraction algorithm, then Kalman filters are used for object tracking. At the same time, SURF key points are detected only in image sections identified as moving...
-
Comparative in vitro research of the Human Aortic Bioprosthesis
PublicationE valuate of the usefulness and reliability of structures based on the analysis of recorded parameters determining the flow through the Human Aortic Bioprosthesis (HAB) have been dealt with. By flow parameters changes determining the performance environment of prosthesis analyzed change of the motion dynamics of the valve leaflets as a function of pressure, thereby determining the degree of alignment of the prosthesis to the performance...
-
THE EFFECT OF WOOD DRYING METHOD ON THE GRANULARITY OF SAWDUST OBTAINED DURING THE SAWING PROCESS USING THE FRAME SAWING MACHINE
PublicationThe experimental results of the study focused on the effect of drying processes of warm air drying at the temperature of 6580°C and warm air-steam mixture drying at the temperature of 105°C of pine and beech wood to the size of sawdust grains created by cutting using RPW 15M frame saw is presented in the paper. Particle size analysis of dry sawdust was performed using two methods - screening method and optical method based on...
-
An Analysis of the Performance of Lightweight CNNs in the Context of Object Detection on Mobile Phones
PublicationConvolutional Neural Networks (CNNs) are widely used in computer vision, which is now increasingly used in mobile phones. The problem is that smartphones do not have much processing power. Initially, CNNs focused solely on increasing accuracy. High-end computing devices are most often used in this type of research. The most popular application of lightweight CNN object detection is real-time image processing, which can be found...
-
Spatially variant PSF modeling in confocal macroscopy
PublicationThe point spread function (PSF) of imaging systems plays an essential role in image reconstruction. In the context of confocal microscopy, optical performance degrades towards the edge of the field of view. In confocal macroscopy, the related artifacts are even stronger, as the field of view is much larger. Because the related PSFs are strongly spatially variant, it is essential to be able to model them with few parameters. The...
-
On the Role of Polarimetric Decomposition and Speckle Filtering Methods for C-Band SAR Wetland Classification Purposes
PublicationPrevious wetlands studies have thoroughly verified the usefulness of data from synthetic aperture radar (SAR) sensors in various acquisition modes. However, the effect of the processing parameters in wetland classification remains poorly explored. In this study, we investigated the influence of speckle filters and decomposition methods with different combinations of filter and decomposition windows sizes on classification accuracy....
-
Combined Single Neuron Unit Activity and Local Field Potential Oscillations in a Human Visual Recognition Memory Task
PublicationGOAL: Activities of neuronal networks range from action potential firing of individual neurons, coordinated oscillations of local neuronal assemblies, and distributed neural populations. Here, we describe recordings using hybrid electrodes, containing both micro- and clinical macroelectrodes, to simultaneously sample both large-scale network oscillations and single neuron spiking activity in the medial temporal lobe structures...
-
Fuzzy rule-based dynamic gesture recognition employing camera & multimedia projector
PublicationIn the paper the system based on camera and multimedia projector enabling a user to control computer applications by dynamic hand gestures is presented. The main objective is to present the gesture recognition methodology which bases on representing hand movement trajectory by motion vectors analyzed using fuzzy rule-based inference. The approach was engineered in the system developed with J2SE and C++ / OpenCV technology. OpenCV...
-
Modeling SPMD Application Execution Time
PublicationParallel applications in a Single Process Multiple Data paradigm assume splitting huge amounts of data to multiple processors working in parallel at small data packets. As the individual data packets are not independent, the processors must interact with each other to exchange results of the calculations with their adjacent partners and take these results into account in their own computations. An example of SPMD is geometric parallelism...
-
Compressible gas density measurement by means of Fourier analysis of interferograms
PublicationThis paper describes a method for nonintrusive compressible gas density measurement by means of automated analysis of interferograms using FFT (Fast Fourier Transform), and its implementation using DFT (Discrete Fourier Transform), that does make this measurement technique a fairly valuable and accessible experimental method. The presented approach makes it possible to use the finite fringe setting of the interferometer, thus reducing...
-
EFFICIENT LINE DETECTION METHOD BASED ON 2D CONVOLUTION FILTER
PublicationThe article proposes an efficient line detection method using a 2D convolution filter. The proposed method was compared with the Hough transform, the most popular method of straight lines detection. The developed method is suitable for local detection of straight lines with a slope from -45˚ to 45˚. Also, it can be used for curve detection which shape is approximated with the short straight sections. The new method is characterized...
-
Interactions with recognized objects
PublicationImplicit interaction combined with object recognition techniques opens a new possibility for gathering data and analyzing user behavior for activity and context recognition. The electronic eyewear platform, eGlasses, is being developed, as an integrated and autonomous system to provide interactions with smart environment. In this paper we present a method for the interactions with the recognized objects that can be used for electronic...
-
Examining Classifiers Applied to Static Hand Gesture Recognition in Novel Sound Mixing System
PublicationThe main objective of the chapter is to present the methodology and results of examining various classifiers (Nearest Neighbor-like algorithm with non-nested generalization (NNge), Naive Bayes, C4.5 (J48), Random Tree, Random Forests, Artificial Neural Networks (Multilayer Perceptron), Support Vector Machine (SVM) used for static gesture recognition. A problem of effective gesture recognition is outlined in the context of the system...
-
Augmented Reality for Privacy-Sensitive Visual Monitoring
PublicationThe paper presents a method for video anonymization and replacing real human silhouettes with virtual 3D figures rendered on the screen. Video stream is processed to detect and to track objects, whereas anonymization stage employs fast blurring method. Substitute 3D figures are animated accordingly to behavior of detected persons. Their location, movement speed, direction, and person height are taken into account during the animation...
-
Night in a big city. Light festivals as a creative medium used at night and their impact on the authority, significance and prestige of a city
PublicationCities in times of global urbanization, are looking for completely new marketing methods, one such tool is the creative use of artificial light after dark, in the form of light festivals. This paper aims to introduce the historical development of the phenomenon and contemporary as well as contemporary illuminations / projects of illuminations of Fete des Lumieres festival in Lyon, Luminale in Frankfurt or on a smaller scale Lights...
-
Mobilenet-V2 Enhanced Parkinson's Disease Prediction with Hybrid Data Integration
PublicationThis study investigates the role of deep learning models, particularly MobileNet-v2, in Parkinson's Disease (PD) detection through handwriting spiral analysis. Handwriting difficulties often signal early signs of PD, necessitating early detection tools due to potential impacts on patients' work capacities. The study utilizes a three-fold approach, including data augmentation, algorithm development for simulated PD image datasets,...
-
Measurement spectrum obtained with the use of ZnO coated microsphere-based fiber-optic sensor - microsphere inspection s.5
Open Research DataApplication of a microsphere-based fiber-optic sensor with 200 nm zinc oxide (ZnO) coating, deposited by Atomic Layer Deposition (ALD) method, for temperature measurements between 100°C and 300°C, is presented. The main advantage of integrating a fiber-optic microsphere with a sensing device is the possibility of monitoring the integrity of the sensor...
-
Measurement spectrum obtained with the use of ZnO coated microsphere-based fiber-optic sensor - microsphere inspection s.4
Open Research DataApplication of a microsphere-based fiber-optic sensor with 200 nm zinc oxide (ZnO) coating, deposited by Atomic Layer Deposition (ALD) method, for temperature measurements between 100°C and 300°C, is presented. The main advantage of integrating a fiber-optic microsphere with a sensing device is the possibility of monitoring the integrity of the sensor...
-
Measurement spectrum obtained with the use of ZnO coated microsphere-based fiber-optic sensor - microsphere inspection s.3
Open Research DataApplication of a microsphere-based fiber-optic sensor with 200 nm zinc oxide (ZnO) coating, deposited by Atomic Layer Deposition (ALD) method, for temperature measurements between 100°C and 300°C, is presented. The main advantage of integrating a fiber-optic microsphere with a sensing device is the possibility of monitoring the integrity of the sensor...
-
Remote Sensing Methods In the Study of the Impact of Long-Term Process of Sulphur Mining on Environmental Changes of the Carpathian Foreland
PublicationThe paper presents research on the extent of impact of sulphur mining process and post-mining activities upon properties of selected elements of the environment, as well as the assessment of the influence of indirect effects resulting from many years' process of exploitation of sulphur deposits in the areas of the Carpathian Foreland (south-east Poland). Within the scope of research conducted, the assessment of the extent of...
-
Floodplain inundation Mapping using SAR Scattering Coefficient Thresholding and Observed Discharge Data
PublicationInundation area time series are important for wetlands monitoring and hydrological model validation. This study is conducted in Biebrza floodplain, which is a natural wetland with complex inundation generation processes. In order to map 2014-2018 series of inundation in the floodplain we test our automatic thresholding method on Sentinel 1 data. The threshold value is optimized using correlation of the inundation area with observed...
-
Visual Detection of People Movement Rules Violation in Crowded Indoor Scenes
PublicationThe paper presents a camera-independent framework for detecting violations of two typical people movement rules that are in force in many public transit terminals: moving in the wrong direction or across designated lanes. Low-level image processing is based on object detection with Gaussian Mixture Models and employs Kalman filters with conflict resolving extensions for the object tracking. In order to allow an effective event...
-
Modeling and analysis of the effectiveness of the guard systemswith dynamic graphs
PublicationIn the following paper it will be presented a new model for analysis (in polynomial time) of the effectiveness of the guard systems. Therewill be presented its practical applications in problems such as searching for the weakest points of the system, planning guards' paths or cameras deployment, switching image from multiple cameras on several monitors, or interception of the intruder. This model is based on describing the guarded...
-
Modified dynamic time warping method applied to handwritten signature authenticity verification
PublicationA signature verification system based on static features and time-domain functions of signals obtained using a tablet has been presented in the paper. The signature verification method, based mainly on dynamic time warping coupled with some signature image features, has been described. The FRR measures reflecting the method’s efficiency have been evaluated for verification attempts performed directly after obtaining model signatures...
-
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...
-
3D Hand Shape Modeling for Automatic Assessing Motor Performance in Parkinson's Disease
PublicationIn this paper a method for hand pattern processing to create a 3D hand model is presented. By applying a complete hand armature to the model obtained, an interpolation of three motor tests for an individual Parkinson's disease patient can be performed. To obtain the 3D hand model the top view of the hand from a web cam is analyzed. The hand contour is examined to find characteristic points that allows for dividing hand image into...
-
Microseismic Monitoring of Hydraulic Fracturing - Data Interpretation Methodology With an Example from Pomerania
PublicationMicroseismic monitoring is a method for localizing fractures induced by hydraulic fracturing in search for shell gas. The data is collected from an array of geophones deployed on the surface or underground. Ground vibrations are recorded and analysed for fracture location, magnitude and breakage mechanism. For successful microseismic monitoring one need a velocity model of underlying formations. The model is further tuned with...
-
Probabilistic sensitivity of limit states of structures : The Monte Carlo simulation
PublicationThe main issue of the paper is the probabilistic sensitivity of the limit states of structures with respect to selected input design variables. Attempt to the problem is done by the dedicated Monte Carlo simulation procedure. Basic design variables are random variables of given probability distributions, presented in the form of random numbers. Uni-parametrical increment of the dominant basic variable (basic variables) is done,...
-
Gesture-based computer control system applied to the interactive whiteboard
PublicationIn the paper the gesture-based computer control system coupled with the dedicated touchless interactive whiteboard is presented. The system engineered enables a user to control any top-most computer application by using one or both hands gestures. First, a review of gesture recognition applications with a focus on methods and algorithms applied is given. Hardware and software solution of the system consisting of a PC, camera, multimedia...
-
Gesture-based computer control system applied to the interactive whiteboard
PublicationIn the paper the gesture-based computer control system coupled with the dedicated touchless interactive whiteboard is presented. The system engineered enables a user to control any top-most computer application by using one or both hands gestures. First, a review of gesture recognition applications with a focus on methods and algorithms applied is given. Hardware and software solution of the system consisting of a PC, camera, multimedia...
-
Adaptive Binarization of Metal Nameplate Images Using the Pixel Voting Approach
PublicationIn the paper, an application of the recently proposed approach to hybrid image binarization based on pixel voting is considered for industrial images. Since such images typically contain the text embossed or engraved in metal nameplates, often non-uniformly illuminated, a proper binarization of such images is usually much harder than for scanned document images, or even for the photos of text documents. Assuming that no single...
-
Landscape perception and the teaching of it in Poland
PublicationA preliminary assessment of the state of knowledge of landscape perception is contained in this article. A survey was carried out of students of architecture and urban planning during the rural design course in the academic year 2017/2018. The pilot project involved the initial establishment of categories of the countryside image. It was assumed that the results represented students’ knowledge of the landscape characteristics and...
-
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,...
-
Zieleń miejska Gdyni okresu międzywojennego oraz problematyka jej zachowania = Urban greenery of Gdynia of interwar period and the issues of its retaining
PublicationURBAN GREENERY OF GDYNIA OF INTERWAR PERIOD AND THE ISSUES OF ITS RETAINING The subject of this article is the urban greenery arrangement in interwar Gdynia and the issues of its retaining after the II World War. The article will concern retaining the authenticity and integrity of the historic urban greenery of Gdynia, both greenery layouts and the plant material creating it. The author of the article will present in what...
-
Matrix-based robust joint fingerprinting and decryption method for multicast distribution of multimedia
PublicationThis paper addresses the problem of unauthorized redistribution of multimedia content by malicious users (pirates). The solution proposed here is a new joint fingerprinting and decryption method which meets the requirements for both imperceptibility and robustness of fingerprints and scalability in terms of design and distribution of fingerprinted multimedia content. The proposed method uses a simple block cipher based on matrix...
-
comparison of circular and rectangular narrow esps with longitudinal wire electrode
PublicationDiesel engines emit fine particles, which are harmful to human and animal health. There are several methods for decrease particulate emission from a diesel engines, but up to now, these methods are not enough effective or very expensive. An electrostatic precipitation was proposed as an alternative method for control of a diesel particulate emission. Therefore, narrow electrostatic precipitators (ESPs) have become a subject of...
-
Hand gesture recognition supported by fuzzy rules and Kalman filters
PublicationThe paper presents a system based on camera and multimediaprojector enabling a user to control computer applications by dynamic hand gestures. Gesture recognition methodology based on representing hand movement trajectory by motion vectors analysed using fuzzy rule-based inference is first given. For effective hand position tracking Kalman filters are employed. The system engineered is developed using J2SE and C++/OpenCV technology....
-
"Shadow" method application within endoscopic examinations of marine engines
PublicationThe paper deals with diagnostic issues concerning endoscopic examinations of the working spaces within marine diesel and gas turbine engines. In the beginning, the endoscopy apparatus being on the laboratory equipment of the Department of Ship Power Plants of Gdansk University of Technology in Poland has been characterized. The endoscopy considerations have been focused on theoretical bases of a digital image processing and especially...
-
"Shadow" method application within endoscopic examinations of marine engines
PublicationThe paper deals with diagnostic issues concerning endoscopic examinations of the working spaces within marine diesel and gas turbine engines. In the beginning, the endoscopy apparatus being on the laboratory equipment of the Department of Ship Power Plants of Gdansk University of Technology in Poland has been characterized. The endoscopy considerations have been focused on theoretical bases of a digital image processing and especially...
-
Efficient signal processing in spectroscopic optical coherence tomography
PublicationSpectroscopic optical coherence tomography (SOCT) is an extension of a standard OCT technique, which allows to obtain depth-resolved, spectroscopic information on the examined sample. It can be used as a source of additional contrast in OCT images e.g. by encoding certain features of the light spectrum into the hue of the image pixels. However, SOCT require computation of time-frequency distributions of each OCT A-scan, what is...
-
Pumping effect measured by PIV method in a multilayer spike electrode EHD device for air cleaning
PublicationDust particles can be harmful for human health when inhaled. Particularly dangerous are submicrometer dust particles, which can contain traces of toxic elements and can easily penetrate into the human respiratory system. Thus, efficient devices for the air cleaning from submicrometer dust particles are needed. Recently, Katatani and Mizuno have proposed an electrohydrody- namic (EHD) device for air cleaning for submicrometer particles....
-
Influence of Toll Collection Method on Motorways on Traffic Safety and Efficiency
PublicationThe paper presents the results of the effectiveness analysis for various forms of toll collection on motorway gates and their impact on safety and efficiency of traffic. The results from field tests that were performed on the A1 motorway in Rusocin were used for the analysis. The research included the analysis of the video image of driver behaviour on the way to the toll booths and at toll booths. The PTV VISSIM tool was used to...
-
Use of radiography images and gray level co-occurrence matrix to investigate gravitational granular flow
PublicationThe paper presents analysis of granular gravitational flow based on radiography images processing. The investigations were conducted for silo model geometry with concentric/eccentric discharging modes. The continuous X-ray radiography scans of granular material distribution, acquired during flow, were obtained by means of an especially designed model silo with rectangular bin and different settings of hopper angles. Image processing...
-
Comparison of the exponential thermal transient parameterization methods with the SMTP method in the unipedicled DIEP flap computer modelling and simulation
PublicationThe aim of this paper is to compare the spatial contrast of the image descriptors obtained via three different thermal transient parameterization methods in Active Dynamic Thermography. The thermal constants and amplitude values of the one- and two- exponential parametrization are compared to the Simplified Magnitude-Temporal Parametrization method (SMTP). The comparison is performed using the data obtained by simulating the cold...
-
Shallow Water Communication with an Object Buried in Bottom Sediments
PublicationUnderwater acoustic communications (UAC) in shallow water applications is a very difficult task. This task becomes even more difficult when there is a need to ensure reliable communication with an object buried in bottom sediments. The article presents a simulation of an acoustic transmission channel in conditions of strong multi-path propagation to an object buried in bottom sediments. The impulse response method was used, supported...
-
Instance segmentation of stack composed of unknown objects
PublicationThe article reviews neural network architectures designed for the segmentation task. It focuses mainly on instance segmentation of stacked objects. The main assumption is that segmentation is based on a color image with an additional depth layer. The paper also introduces the Stacked Bricks Dataset based on three cameras: RealSense L515, ZED2, and a synthetic one. Selected architectures: DeepLab, Mask RCNN, DEtection TRansformer,...