Search
Search results for: OBJECT DETECTION, BACKGROUND SUBTRACTION
-
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,...
-
Sea ice floe size and shape data from a very high resolution satellite image (Knox Coast, East Antarctica)
Open Research DataThis dataset contains floe size distribution data from a very high resolution (pixel size: 0.3 m) optical satellite image of sea ice, acquired on 16. Feb. 2019 off the Knox Coast (East Antarctica). The image shows relatively small ice floes produced by wave-induced breakup of landfast ice between the Mill Island and Bowman Island. The ice floes are...
-
The circle object detection with the use of Msplit estimation
PublicationThe paper presents the use of Msplit(q) - estimation in the filtration and aggregation of point clouds containing a known number of elliptical shapes with preliminary unknown - locations and dimensions. These theoretical solutions may have practical relevance especially in the modelling of terrestrial laser scanning data of objects that have similar shape to circles. Mentioned shapes can be scanned of tree trunks, columns, gutters,...
-
Point-pixel fusion for object detection and depth estimation
Publication -
Multiscaled Hybrid Features Generation for AdaBoost Object Detection
PublicationThis work presents the multiscaled version of modified census features in graphical objects detection with AdaBoost cascade training algorithm. Several experiments with face detector training process demonstrate better performance of such features over ordinal census and Haar-like approaches. The possibilities to join multiscaled census and Haar features in single hybrid cascade of strong classifiers are also elaborated and tested....
-
Magnetic field maps of an astable multivibrator in frequency range from 100 kHz to 50 MHz
Open Research DataThe data presents a result of near field measurements of electromagnetic emissions radiated from the PCB of a small electronic device. An efficient method of modelling the magnetic and electric field emissions is the measurements in the near field using electric and magnetic probes. The attached files contain magnetic field maps created on based measurements...
-
Electric field maps of an astable multivibrator in frequency range from 30 MHz to 3 GHz
Open Research DataThe data presents a result of near field measurements of electromagnetic emissions radiated from the PCB of a small electronic device. An efficient method of modelling the magnetic and electric field emissions is the measurements in the near field using electric and magnetic probes. The attached files contain electric field maps created on based measurements...
-
Resolving conflicts in object tracking for automatic detection of events in video
PublicationW referacie przedstawiono algorytm rozwiązywania konfliktów w śledzeniu obiektów ruchomych. Proponowana metoda wykorzystuje predykcję stanu obiektu obliczaną przez filtry Kalmana oraz dopasowuje wykryte obiekty do struktur śledzących ich ruch na podstawie deskryptorów koloru i tekstury. Omówiono specyficzne sytuacje powodujące konflikty, takie jak rozdzielanie obiektów. Przedstawiono wyniki testów. Algorytm może być zastosowany...
-
Resolving conflicts in object tracking for automatic detection of events in video
PublicationAn algorithm for resolving conflicts in tracking of moving objects is presented. The proposed approach utilizes predicted states calculated by Kalman filters for estimation of trackers position, then it uses color and texture descriptors in order to match moving objects with trackers. Problematic situations, such as splitting objects, are addressed. Test results are presented and discussed. The algorithm may be used in the system...
-
Magnetic field maps of an astable multivibrator in frequency range from 30 MHz to 3 GHz – selective detection
Open Research DataThe data presents a result of near field measurements of electromagnetic emissions radiated from the PCB of a small electronic device. An efficient method of modelling the magnetic and electric field emissions is the measurements in the near field using electric and magnetic probes. The attached files contain magnetic field maps created on based measurements...
-
Modular machine learning system for training object detection algorithms on a supercomputer
PublicationW pracy zaprezentowano architekturę systemu służącego do tworzenia algorytmów wykorzystujących metodę AdaBoost i służących do wykrywania obiektów (np. twarzy) na obrazach. System został podzielony na wyspecjalizowane moduły w celu umożliwienia łatwej rozbudowy i efektywnego zrównoleglenia implementacji przeznaczonej dla superkomputera. Na przykład, system może być rozszerzony o nowe cechy i algorytmy ich ekstrakcji bez konieczności...
-
Aleksandra Karpus dr inż.
PeopleAleksandra Karpus jest absolwentką Matematyki Stosowanej na Wydziale Fizyki Technicznej i Matematyki Stosowanej oraz Informatyki na Wydziale Elektroniki, Telekomunikacji i Informatyki Politechniki Gdańskiej. W latach 2011-2014 pracowała z danymi w przemyśle, wykorzystując bazy danych Oracle. Od 2014 roku jest zawodowo związana z Politechniką Gdańską, obecnie jest zatrudniona na stanowisku adiunkta naukowo-dydaktycznego w Katedrze...
-
Object-oriented approach to oil spill detection using ENVISAT ASAR images
Publication -
An object-based SAR image iceberg detection algorithm applied to the Amundsen Sea
Publication -
City as a product. Architecture as an Economic Instrument. Are Global Cities People-Friendly Places?
PublicationWhile spending time in our everyday urban environment do we ever think how particular architecture influences the economic value of that space? Space has its measurable financial value. From the economic point of view a place can be treated as a product that fights for appearing in tourists’ and investors’ consciousnesses. Space - treated as an object of demand and supply - becomes an element in a marketing game. To reach its...
-
Exploring the impact of cultural context on eye-tracking studies of architectural monuments in selected European cities: Sustainable heritage management.
PublicationSustainable management of architectural heritage requires conducting an inclusive diagnosis of users’ opinions, considering both residents and tourists as the recipients of urban space. Given the cultural diversity within these groups, proposing the use of eye-trackers (ET) as an alternative to traditional public consultation prompts the need to assess the method’s advantages and disadvantages. It remains uncertain whether individuals...
-
Marta Rusnak, Marta Szmigiel, Malwina Geniusz, Zofia Koszewicz, Monika Magdziak-Tokłowicz, Exploring the impact of cultural context on eye-tracking studies of architectural monuments in selected European cities: Sustainable heritage management.
PublicationSustainable management of architectural heritage requires conducting an inclusive diagnosis of users' opinions, considering both residents and tourists as the recipients of urban space. Given the cultural diversity within these groups, proposing the use of eye-trackers (ET) as an alternative to traditional public consultation prompts the need to assess the method's advantages and disadvantages. It remains uncertain whether individuals...
-
TDOA versus ATDOA for wide area multilateration system
PublicationThis paper outlines a new method of a location service (LCS) in the asynchronous wireless networks (AWNs) where the nodes (base stations) operate asynchronously in relation to one another. This method, called asynchronous time difference of arrival (ATDOA), enables the calculation of the position of the mobile object (MO) through the measurements taken by a set of non-synchronized fixed nodes and is based on the measurement of...
-
Mirosław Wołoszyn dr hab. inż.
PeopleMirosław Wołoszyn born in 1963 in Gdynia. He received the M.Sc. degree in 1987, the Ph.D. degree in 1997, and the D.Sc. (‘habilitation’) degree in 2013, all from the Gdańsk University of Technology. Since 1987 he has been with the above university, where he is currently Associate Professor of Electrical Engineering. His research interests include localization and identification of ferromagnetic objects by means of the magnetometric...
-
Moving object tracking algorithm evaluation in autonomous surveillance system
PublicationResults of evaluation of video object tracking algorithm being a part of an autonomous surveillance system are presented. The algorithm was investigated employing a set of benchmarks recorded locally. The precision of object detection, evaluated with such metrics as fragmentation, object area recall and object precision, is in focus. The experiments aimed at examining the detection quality using various object detection algorithm...
-
A model of stealth maritime object having some innovative solutions concerning the object form, structure and materials.
Open Research DataThe aim of the project is to work out a model of the stealth maritime object which will have innovative solutions concerning the object form, structure and materials. These solutions should enable a modification of combinations of the object features defining the object stealth characteristics (difficulty of the object detection in the water). It is...
-
Automatic detection of abandoned luggage employing a dual camera system
PublicationA system for automatic detection of events using a system of fixed and PTZ (pan-tilt-zoom) cameras is described. Images from the fixed camera are analyzed by means of object detection and tracking. Event detection system uses a set of rules to analyze data on the tracked moving objects and to detect defined events. A PTZ camera is used to obtain a detailed view of a selected object. A procedure for conversion between the pixel...
-
Detection and segmentation of moving vehicles and trains using Gaussian mixtures, shadow detection and morphological processing
PublicationSolution presented in this paper combines background modelling, shadow detection and morphological and temporal processing into one system responsible for detection and segmentation of moving objects recorded with a static camera. Vehicles and trains are detected based on their pixellevel difference from the continually updated background model utilizing a Gaussian mixture calculated separately for every pixel. The shadow detection...
-
Objects classification based on their physical sizes for detection of events in camera images
PublicationIn the paper, a method of estimation of the physical sizes of the objects tracked in the video surveillance system, and a simple module for object classification based on the estimated physical sizes, are presented. The results of object classification are then used for automatic detection of various types of events in the camera image.
-
Comparison of edge detection algorithms for electric wire recognition
PublicationEdge detection is the preliminary step in image processing for object detection and recognition procedure. It allows to remove useless information and reduce amount of data before further analysis. The paper contains the comparison of edge detection algorithms optimized for detection of horizontal edges. For comparison purposes the algorithms were implemented in the developed application dedicated to detection of electric line...
-
Bartosz Szostak mgr inż.
PeopleBartosz Szostak graduated with a degree in engineering, specializing in Geodesy and Cartography, at the Gdansk University of Technology in 2019. On 2021, he graduated with a Master's degree also in the field of Geodesy and Cartography at the Gdansk University of Technology. The topics covered in his thesis were machine learning and object detection.
-
Hierarchical 2-step neural-based LEGO bricks detection and labeling
PublicationLEGO bricks are extremely popular and allow the creation of almost any type of construction due to multiple shapes available. LEGO building requires however proper brick arrangement, usually done by shape. With over 3700 different LEGO parts this can be troublesome. In this paper, we propose a solution for object detection and annotation on images. The solution is designed as a part of an automated LEGO bricks arrangement. The...
-
Identification of regions of interest in video for a traffic monitoring system
PublicationA system for automatic event detection in the camera image is presented in this paper. A method of limiting a region of interest to relevant parts of the image using a set of processing procedures is proposed. Image processing includes object detection with shadow removal performed in the modified YCbCr color space instead of RGB. The proposed procedures help to reduce the complexity of image processing algorithm and result in...
-
Performance evaluation of GPS anti-spoofing system based on antenna array processing
PublicationThis article concerns the problem of detection and mitigation of spoofing attacks in Global Navigation Satellite Systems. First, background information about spoofing and its possible countermeasures is provided. Next, concept of a novel GPS anti-spoofing system based on antenna array processing is presented and implementation of prototype of this system is described. The third section outlines the performance evaluation of proposed...
-
Implementation of control system and tracking objects in a Quadcopter
PublicationIn this paper, we implement a quadcopter assembly with control and navigation module. The project also includes the design of the control panel for the operator which consists of a set of the micro-controller and the glove equipped with sensors and buttons. The panel has a touch screen which displays current parameters such as vehicle status, including information about orientation and geographical coordinates. The concept of quadcopter...
-
Extraction of stable foreground image regions for unattended luggage detection
PublicationA novel approach to detection of stationary objects in the video stream is presented. Stationary objects are these separated from the static background, but remaining motionless for a prolonged time. Extraction of stationary objects from images is useful in automatic detection of unattended luggage. The proposed algorithm is based on detection of image regions containing foreground image pixels having stable values in time and...
-
A novel concept for tissue-metal detection and differentiation using an inductive proximity sensor
PublicationIn this paper a novel application of inductive proximity sensors for detection of living tissue by means of measurements of the coil impedance changes at different frequencies is described. The mathematical analyses utilizing Bessel function estimation include detected object size and its distance from a sensor. The main aim of this study is to prove the possibility of distinguishing between metal objects and living tissues. The...
-
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...
-
Optimizing of target detection and tracking processes realized on consoles of passive sonar with linear towed antenna
PublicationThe long-range passive towed sonar was first modernised a few years ago. Building on operator experience over that period, a concept was developed of optimising the tasks performed by sonar operators, and improving forms of imaging to inform about object detection and support object tracking. The concept was implemented and successfully tested during ships’ manoeuvres. The optimisation of operator tasks was designed to keep listening...
-
Pomiarowa ocena efektywności systemu antyspoofingowego opartego na przetwarzaniu wieloantenowym
PublicationNiniejszy referat dotyczy problemu wykrywania i przeciwdziałania atakom typu spoofing w globalnych systemach nawigacji satelitarnej. Pierwsza część referatu stanowi wprowadzenie w temat spoofingu oraz przegląd sposobów jemu przeciwdziałania. Następnie, przedstawiono autorską koncepcję systemu antyspoofingowego opartego na przetwarzaniu wieloantenowym, a także opisano implementację prototypu tego systemu. Część trzecią poświęcono...
-
Cartographic Representation of Route Reconstruction Results in Video Surveillance System
PublicationThe video streams available in a surveillance system distributed on the wide area may be accompanied by metadata are obtained as a result of video processing. Many algorithms applied to surveillance systems, e.g. event detection or object tracking, are strictly connected with localization of the object and reconstruction of its route. Drawing related information on a plan of a building or on a map of the city can facilitate the...
-
Multi-Stage Video Analysis Framework
PublicationThe chapter is organized as follows. Section 2 presents the general structure of the proposed framework and a method of data exchange between system elements. Section 3 is describing the low-level analysis modules for detection and tracking of moving objects. In Section 4 we present the object classification module. Sections 5 and 6 describe specialized modules for detection and recognition of faces and license plates, respectively....
-
Tracking Moving Objects in Video Surveillance Systems with Kalman and Particle Filters – A Practical Approach
PublicationThis Chapter focuses on the first type of object tracking algorithms, namely on Kalman and particle filters. A theory of these algorithms may be found in many publications, there are also reports on implementation of these approaches to object tracking in video. However, developers of VCA systems still face two important problems. The first one is related to obtaining accurate measurements of positions and sizes of the tracked...
-
A New Adaptive Method for the Extraction of Steel Design Structures from an Integrated Point Cloud
Open Research DataA new automatic and adaptive algorithm for edge extraction from a random point cloud was developed and presented herein. The proposed algorithm was tested using real measurement data. The developed algorithm is able to realistically reduce the amount of redundant data and correctly extract stable edges representing the geometric structures of a studied...
-
Diagnostics of high-voltage varistor quality using resonant ultrasound spectroscopy
PublicationResonant spectroscopy is an effective nondestructive testing method focused on defects detection in tested objects. Resonance spectrum contains an information about the mechanical properties and particularly about the structural homogeneity of the whole object. A swept frequency method has been used to measure the selected resonant frequencies range. The system for measurement of resonant ultrasound spectroscopy of high-voltage...
-
Naval mine detection system based of FPGA circuit
PublicationElectrochemical processes take place in a metal object immersed into sea water even if an anticorrosive coating is applied [1]. As a result, flowing field appears around the object. There are naval mines between many other objects situated in the sea. Naval mines can be put in the seabed in order to be more difficult to detect by sonars. Such a mine is located on the line demarking two environments of different electrical conductivity....
-
Bioterrorism — characteristics and possibilities of prevention
PublicationIn the paper bioterrorist threats have been presented. Historical background and possible methods of attacks have been described. The most dangerous pathogens and disease entities have been classified. Selected methods of detection and identification of biological weapon have been presented. The wireless system for threats monitoring — developed at Gdansk University of Technology — has been described.
-
Noise sources in Raman spectroscopy of biological objects
PublicationWe present an overview of noise sources deteriorating the quality of the recorded biological Raman spectra and the ability to determine the specimen composition. The acquired Raman spectra exhibit intense additive noise components or drifts because of low intensity of the scattered light. Therefore we have to apply expensive or bulky measurement setups to limit their inherent noise or to apply additional signal processing to reduce...
-
Distance measurement errors in silent FM-CW sonar with matched filtering
PublicationThe secretiveness of sonar operation can be achieved by using continuous frequency-modulated sounding signalswith reduced power and significantly prolonged repeat time. The application of matched filtration in the sonarreceiver provides optimal conditions for detection against the background of white noise and reverberation, and avery good resolution of distance measurements of motionless targets. The article shows that target...
-
Polarization-sensitive optical coherence tomography for ceramic materials inspection
PublicationCeramics production is looking for a fast, reliable and non-destructive method that can be implemented on site for defect detection and analysis. In this paper we present polarization-sensitive optical coherence tomography (PS-OCT) as a method for defect inspection. Proposed extensions to standard OCT provide additional information for complete characterization of tested object. We compare OCT and microscope imaging that can easily...
-
Video Semantic Analysis Framework based on Run-time Production Rules - Towards Cognitive Vision
PublicationThis paper proposes a service-oriented architecture for video analysis which separates object detection from event recognition. Our aim is to introduce new tools to be considered in the pathway towards Cognitive Vision as a support for classical Computer Vision techniques that have been broadly used by the scientific community. In the article, we particularly focus in solving some of the reported scalability issues found in current...
-
Elżbieta Marczak dr inż. arch.
PeopleElżbieta Marczak (Architect, Phd) – Assistant Professor at the Faculty of Architecture, Gdańsk University of Technology, Department of Urban Architecture and Waterscapes (since 1997). Studies at the Faculty of Architecture GUT graduated in 1995. Degree of Doctor obtained in 2006 presenting dissertation The ship’s architecture against the background of the maritime transport development. Her research and publications focus on...
-
The Innovative Faculty for Innovative Technologies
PublicationA leaflet describing Faculty of Electronics, Telecommunications and Informatics, Gdańsk University of Technology. Multimedia Systems Department described laboratories and prototypes of: Auditory-visual attention stimulator, Automatic video event detection, Object re-identification application for multi-camera surveillance systems, Object Tracking and Automatic Master-Slave PTZ Camera Positioning System, Passive Acoustic Radar,...
-
Toward Robust Pedestrian Detection With Data Augmentation
PublicationIn this article, the problem of creating a safe pedestrian detection model that can operate in the real world is tackled. While recent advances have led to significantly improved detection accuracy on various benchmarks, existing deep learning models are vulnerable to invisible to the human eye changes in the input image which raises concerns about its safety. A popular and simple technique for improving robustness is using data...
-
Performance Analysis of Interaction between Smart Glasses and Smart Objects Using Image-Based Object Identification
PublicationWe propose the use of smart glasses to collaborate with smart objects in the Internet of Things environment. Particularly we are focusing on new interaction methods and the analysis of acceptable reaction times in the process of object recognition using smart glasses. We evaluated the proposed method using user studies and experiments with three different smart glasses: Google Glass, Epson Moverio, and the developed eGlasses platform....