Filtry
wszystkich: 82
Wyniki wyszukiwania dla: SEGMENTATION BY PART
-
Abdominal Aortic Aneurysm segmentation from contrast-enhanced computed tomography angiography using deep convolutional networks
PublikacjaOne of the most common imaging methods for diagnosing an abdominal aortic aneurysm, and an endoleak detection is computed tomography angiography. In this paper, we address the problem of aorta and thrombus semantic segmentation, what is a mandatory step to estimate aortic aneurysm diameter. Three end-to-end convolutional neural networks were trained and evaluated. Finally, we proposed an ensemble of deep neural networks with underlying...
-
Segmentation of academic community for the purposes of mobility plan development – case study of Gdansk University of Technology
PublikacjaThe objective of the paper is to analyse the structure of academic community for its transport behaviour and attitudes using the example of the Gdansk University of Technology (the GUT) in Poland. Once understood, the group can be divided into homogenous sub-groups and studied for their potential and ways to influence their behaviour, attitudes and transport patterns. The results may be used to develop dedicated actions designed...
-
Three-dimensional modeling and automatic analysis of the human nasal cavity and paranasal sinuses using the computational fluid dynamics method
PublikacjaPurpose The goal of this study was to develop a complete workflow allowing for conducting computational fluid dynam- ics (CFD) simulation of airflow through the upper airways based on computed tomography (CT) and cone-beam computed tomography (CBCT) studies of individual adult patients. Methods This study is based on CT images of 16 patients. Image processing and model generation of the human nasal cavity and paranasal sinuses...
-
Identification of Emotions Based on Human Facial Expressions Using a Color-Space Approach
PublikacjaHCI technology improves human-computer interaction. Such communication can be carried out with the use of emotions that are visible on the human face since birth. In this paper the Emotion system for detecting and recognizing facial expressions, developed in the MSc work, is presented. The system recognizes emotion from webcam video in real time. It is based on color segmentation and morphological operations. The system uses a...
-
Pipelined division of signed numbers with the use of residue arithmetic in FPGA
PublikacjaAn architecture of a pipelined signed residue divider for small number ranges is presented. The divider makes use of the multiplicative division algorithm where initially the reciprocal of the divisor is calculated and subsequently multiplied by the dividend. The divisor represented in the signed binary form is used to compute the approximated reciprocal in the residue form by the table look-up. In order to reduce the needed length...
-
Detection and segmentation of moving vehicles and trains using Gaussian mixtures, shadow detection and morphological processing
PublikacjaSolution 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...
-
Investigations of speech signal parameters with regard to articulation influences
PublikacjaW pracy zostało podjęte zagadnienie parametryzacji sygnału mowy w kontekście ekstrakcji cech biometrycznych. Analizowane parametry to parametry cepstralne (cepstrum liniowe i mel-cepstrum, czyli MFCC), parametry liniowej predykcji (LPC) oraz momenty widmowe i parametr F0. Zastosowano analize w krótkich stałych segmentach sygnału z zastosowaniem dużego zakładkowania, tzw. ''implicite segmentation''. Umożliwiło to zaobserwowanie...
-
An Overview of Image Analysis Techniques in Endoscopic Bleeding Detection
PublikacjaAuthors review the existing bleeding detection methods focusing their attention on the image processing techniques utilised in the algorithms. In the article, 18 methods were analysed and their functional components were identified. The authors proposed six different groups, to which algorithms’ components were assigned: colour techniques, reflecting features of pixels as individual values, texture techniques, considering spatial...
-
MSIS sonar image segmentation method based on underwater viewshed analysis and high-density seabed model
PublikacjaHigh resolution images of Mechanically Scanned Imaging Sonars can bring detailed representation of underwater area if favorable conditions for acoustic signal to propagate are provided. However to properly asses underwater situation based solely on such data can be challenging for less than proficient interpreter. In this paper we propose a method to enhance interpretative potential of MSIS image by dividing it in to subareas depending...
-
Vehicle detector training with minimal supervision
PublikacjaRecently many efficient object detectors based on convolutional neural networks (CNN) have been developed and they achieved impressive performance on many computer vision tasks. However, in order to achieve practical results, CNNs require really large annotated datasets for training. While many such databases are available, many of them can only be used for research purposes. Also some problems exist where such datasets are not...
-
Improving methods to calculate the loss of ecosystem services provided by urban trees using LiDAR and aerial orthophotos
PublikacjaIn this paper we propose a methodology for combining remotely sensed data with field measurements to assess selected tree parameters (diameter at breast height (DBH) and tree species) required by the i-Tree Eco model to estimate ecosystem services (ES) provided by urban trees. We determined values of ES provided by trees in 2017 in Racibórz (a city in South Poland) and estimated the loss of ES from January 1, 2017 to March 5, 2017,...
-
Impact of Shifting Time-Window Post-Processing on the Quality of Face Detection Algorithms
PublikacjaWe consider binary classification algorithms, which operate on single frames from video sequences. Such a class of algorithms is named OFA (One Frame Analyzed). Two such algorithms for facial detection are compared in terms of their susceptibility to the FSA (Frame Sequence Analysis) method. It introduces a shifting time-window improvement, which includes the temporal context of frames in a post-processing step that improves the...
-
Rapid multi-objective design optimisation of compact microwave couplers by means of physics-based surrogates
PublikacjaThe authors introduce a methodology for fast multi-objective design optimisation of miniaturised microwave couplers. The approach exploits the surrogate-based optimisation paradigm with an underlying low-fidelity model constructed from an equivalent circuit of the structure under consideration, corrected through implicit and frequency space mapping. A fast prediction tool obtained this way is subsequently optimised by a multi-objective...
-
Controlling computer by lip gestures employing neural network
PublikacjaResults of experiments regarding lip gesture recognition with an artificial neural network are discussed. The neural network module forms the core element of a multimodal human-computer interface called LipMouse. This solution allows a user to work on a computer using lip movements and gestures. A user face is detected in a video stream from a standard web camera using a cascade of boosted classifiers working with Haar-like features....
-
C2 NIWA Community-Segmentation Criteria and Building Brand Associations on the Example of a Selected Target Group
PublikacjaEvery organization which offers products or services wishes to communicate with their customers in the most effective way. This kind of communication is based on proper selection of target groups, which are extracted in the process of market segmentation. That is way it is very important to ask the question to whom the message is to be directed and what kind of message we want to give. This article describes the selection criteria...
-
Hazard Control in Industrial Environments: A Knowledge-Vision-Based Approach
PublikacjaThis paper proposes the integration of image processing techniques (such as image segmentation, feature extraction and selection) and a knowledge representation approach in a framework for the development of an automatic system able to identify, in real time, unsafe activities in industrial environments. In this framework, the visual information (feature extraction) acquired from video-camera images and other context based gathered...
-
Endoscopic Video Classification with the Consideration of Temporal Patterns
PublikacjaThe article describes a novel approach to automatic recognition and classification of diseases in endoscopic videos. Current directions of research in this field are discussed. Most presented methods focus on processing single frames and do not take into consideration the temporal relationship between continuous classifications. Existing approaches that consider the temporal structure of an incoming frame sequence are focused on...
-
MRI-Derived Subcutaneous and Visceral Adipose Tissue Reference Values for Children Aged 6 to Under 18 Years
PublikacjaThe assessment of body composition in pediatric population is essential for proper nutritional support during hospitalization. However, currently available methods have limitations. This study aims to propose a novel approach for nutrition status assessment and introduce magnetic resonance imaging (MRI)-derived subcutaneous and visceral fat normative reference values. A total of 262 healthy subjects aged from 6 to 18 years underwent...
-
Robust Object Detection with Multi-input Multi-output Faster R-CNN
PublikacjaRecent 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,...
-
Robust Object Detection with Multi-input Multi-output Faster R-CNN
PublikacjaRecent 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,...
-
Workforce mobility against the background of labour market duality theory – the example of selected OECD countries
PublikacjaThe paper aims to present an empirical study of labour market segmentation (LMS) hypothesis. According to the dual labour market theory jobs can be divided into two groups: primary and secondary jobs, with enter barriers into the first one. The primary jobs are usually described with relative high wages, whereas secondary jobs provide lower level of wages. In this paper we first examine the main sectors (according to the ISIC rev....
-
Comparison of Classification Methods for EEG Signals of Real and Imaginary Motion
PublikacjaThe classification of EEG signals provides an important element of brain-computer interface (BCI) applications, underlying an efficient interaction between a human and a computer application. The BCI applications can be especially useful for people with disabilities. Numerous experiments aim at recognition of motion intent of left or right hand being useful for locked-in-state or paralyzed subjects in controlling computer applications....
-
Closer Look at the Uncertainty Estimation in Semantic Segmentation under Distributional Shift
PublikacjaWhile recent computer vision algorithms achieve impressive performance on many benchmarks, they lack robustness - presented with an image from a different distribution, (e.g. weather or lighting conditions not considered during training), they may produce an erroneous prediction. Therefore, it is desired that such a model will be able to reliably predict its confidence measure. In this work, uncertainty estimation for the task...
-
Identification of Emotional States Using Phantom Miro M310 Camera
PublikacjaThe purpose of this paper is to present the possibilities associated with the use of remote sensing methods in identifying human emotional states, and to present the results of the research conducted by the authors in this field. The studies presented involved the use of advanced image analysis to identify areas on the human face that change their activity along with emotional expression. Most of the research carried out in laboratories...
-
Detection of Alzheimer's disease using Otsu thresholding with tunicate swarm algorithm and deep belief network
PublikacjaIntroduction: Alzheimer’s Disease (AD) is a degenerative brain disorder characterized by cognitive and memory dysfunctions. The early detection of AD is necessary to reduce the mortality rate through slowing down its progression. The prevention and detection of AD is the emerging research topic for many researchers. The structural Magnetic Resonance Imaging (sMRI) is an extensively used imaging technique in detection of AD, because...
-
Melanoma skin cancer detection using mask-RCNN with modified GRU model
PublikacjaIntroduction: Melanoma Skin Cancer (MSC) is a type of cancer in the human body; therefore, early disease diagnosis is essential for reducing the mortality rate. However, dermoscopic image analysis poses challenges due to factors such as color illumination, light reflections, and the varying sizes and shapes of lesions. To overcome these challenges, an automated framework is proposed in this manuscript. Methods: Initially, dermoscopic...
-
SkinDepth - synthetic 3D skin lesion database
Dane BadawczeSkinDepth is the first synthetic 3D skin lesion database. The release of SkinDepth dataset intends to contribute to the development of algorithms for:
-
Video content analysis in the urban area telemonitoring system
PublikacjaThe task of constant monitoring of video streams from a large number of cameras and reviewing the recordings in order to find a specified event requires a considerable amount of time and effort from the system operators and it is prone to errors. A solution to this problem is an automatic system for constant analysis of camera images being able to raise an alarm if a predefined event is detected. The chapter presents various aspects...
-
CMGNet: Context-aware middle-layer guidance network for salient object detection
PublikacjaSalient object detection (SOD) is a critical task in computer vision that involves accurately identifying and segmenting visually significant objects in an image. To address the challenges of gridding issues and feature...
-
Vident-real: an intra-oral video dataset for multi-task learning
Dane BadawczeWe introduce Vident-real, a large dataset of 100 video sequences of intra-oral scenes from real conservative dental treatments performed at the Medical University of Gdańsk, Poland. The dataset can be used for multi-task learning methods including:
-
Automatic Threat Detection for Historic Buildings in Dark Places Based on the Modified OptD Method
PublikacjaHistoric buildings, due to their architectural, cultural, and historical value, are the subject of preservation and conservatory works. Such operations are preceded by an inventory of the object. One of the tools that can be applied for such purposes is Light Detection and Ranging (LiDAR). This technology provides information about the position, reflection, and intensity values of individual points; thus, it allows for the creation...
-
Hardware-Software Implementation of a Sensor Network for CityTraffic Monitoring Using the FPGA- and ASIC-Based Sensor Nodes
PublikacjaArtykuł opisuje prototypową sieć sensorową do monitorowania ruchu pojazdów w mieście. Węzły sieci sensorowej, wyposażone w kamerę o niskiej rozdzielczości, obserwują ulice i wykrywają poruszające się obiekty. Detekcja obiektów jest realizowana w oparciu o własny algorytm segmentacji obrazów, wykorzystujący podwójne odejmowanie tła, wykrywanie krawędzi i cieni, działający na dedykowanym systemie mikroelektronicznym typu ''System...