Wyszukiwarka
Wyniki wyszukiwania dla: images analysis
-
Reduction of parasitic pitch variations in archival musical recordings
PublikacjaA new method for reducing parasitic pitch variations in archival audio recordings is presented. The method is intended for analyzing movie soundtracks recorded in optical films. It utilizes image processing for calculating and reducing effects of tape shrinkage being one of the main reasons for parasitic pitch variations in audio accompanying moving images. As long as the film tape characteristics are known the new method can be...
-
Automatic detection of abandoned luggage employing a dual camera system
PublikacjaA 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...
-
CMOS realisation of analogue processor for early vision processing
PublikacjaThe architecture concept of a high-speed low-power analogue vision chip, which performs low-level real-time image algorithms ispresented. The proof-of-concept prototype vision chip containing 32 × 32 photosensor array and 32 analogue processors is fabricated usinga 0.35 μm CMOS technology. The prototype can be configured to register and process images with very high speed, reaching 2000 framesper second, or achieve very low power...
-
3D-Breast System for Determining the Volume of Tissue Needed for Breast Reconstruction
Publikacja3D imaging systems can be used to effectively determine breast volumes for surgical applications. This article presents methods for surface reconstruction and volume determination based on the point cloud created by 3D imaging. Such a system would be used to accurately estimate breast volume in patients classified for breast reconstruction surgery at plastic surgery centers. To develop such a system, various methods of determining...
-
Optimization of the System for Determining the Volume of Tissue Needed for Breast Reconstruction
PublikacjaThis article presents techniques for reconstructing surfaces and volume calculations using a point cloud generated from 3D imaging. The main objective of this article was to optimize the voxel size for the most accurate representation of the surface of the female breast. We experimented with different methods for determining volume using images from the Intel D435i camera. In addition, we designed application and measurement station...
-
Reconstruction of 3D image of corona discharge streamer
PublikacjaIn this paper, the method of reconstruction of the 3D structure of streamers in DC positive corona discharge in nozzle-to-plate electrode configuration is presented. For reconstructing of 3D image of corona discharge streamer we propose a stereographical method, where streamers are observed from several directions simultaneously. The multi-directional observation enabled to obtain fine positional coordinates of streamers for a...
-
Training of Deep Learning Models Using Synthetic Datasets
PublikacjaIn 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...
-
Fusion-based Representation Learning Model for Multimode User-generated Social Network Content
PublikacjaAs mobile networks and APPs are developed, user-generated content (UGC), which includes multi-source heterogeneous data like user reviews, tags, scores, images, and videos, has become an essential basis for improving the quality of personalized services. Due to the multi-source heterogeneous nature of the data, big data fusion offers both promise and drawbacks. With the rise of mobile networks and applications, UGC, which includes...
-
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...
-
DEEP CONVOLUTIONAL NEURAL NETWORKS AS A DECISION SUPPORT TOOL IN MEDICAL PROBLEMS – MALIGNANT MELANOMA CASE STUDY
PublikacjaThe paper presents utilization of one of the latest tool from the group of Machine learning techniques, namely Deep Convolutional Neural Networks (CNN), in process of decision making in selected medical problems. After the survey of the most successful applications of CNN in solving medical problems, the paper focuses on the very difficult problem of automatic analyses of the skin lesions. The authors propose the CNN structure...
-
Assessment of copper surface coverage with corrosion inhibitor using AFM-based local electrical measurements
PublikacjaThe paper presents a new method of assessment of metal surface coverage with corrosion inhibitor and thus of inhibitor protective performance. It is based on the atomic force microscopy measurement performed in a contact mode. Apart from topography images the proposed approach allows acquisition of local DC maps and local electrical impedance spectra via application of DC bias voltage or AC perturbation signal between the conductive...
-
Active Dynamic Thermography imaging of wound healing processes in cardio surgery
PublikacjaThe surgery is a branch of medicine, that is integrally connected to wounds. Despite using sterile tools and compliance with aseptic rolls, some of the surgery wounds become infected. In clinical practice there is a lack of cheap, objective methods and tools for quantitative definition and estimation of the surgery wound healing progress. This paper presents preliminary results of Active Dynamic Thermography (ADT) parametric imaging...
-
"3D-Breast System for Determining the Volume of Tissue Needed for Breast Reconstruction"
PublikacjaThis article presents methods for surface reconstruction and volume determination based on the point cloud created by 3D imaging. Such a system would be used to accurately estimate breast volume in patients classified for breast reconstruction surgery at plastic surgery centers. To develop such a system, various methods of determining volume, based on images from the Intel D435i camera, were tested. In addition, an application...
-
Evaluating Performance and Accuracy Improvements for Attention-OCR
PublikacjaIn this paper we evaluated a set of potential improvements to the successful Attention-OCR architecture, designed to predict multiline text from unconstrained scenes in real-world images. We investigated the impact of several optimizations on model’s accuracy, including employing dynamic RNNs (Recurrent Neural Networks), scheduled sampling, BiLSTM (Bidirectional Long Short-Term Memory) and a modified attention model. BiLSTM was...
-
Visual perception of vowels from static and dynamic cues
PublikacjaThe purpose of the study was to analyse human identification of Polish vowels from static and dynamic durationally slowed visual cues. A total of 152 participants identified 6 Polish vowels produced by 4 speakers from static (still images) and dynamic (videos) cues. The results show that 59% of static vowels and 63% of dynamic vowels were successfully identified. There was a strong confusion between vowels within front, central,...
-
ISSUES OF CLASSIFICATION FUNCTION CONTINUITY IN ENDOSCOPIC VIDEO CLASSIFICATION
PublikacjaIn the article a new way of analyzing the properties of feature vector functions (FVF) and classiers of images in a video stream is proposed. The general idea is based on focusing of the perceived continuity of the FVF and classier functions. Issues related to creating an exact mathematical model are discussed and a simplied solution is proposed. An exemplary algorithm is evaluated on three exemplary video sequences. The acquired...
-
Comparison of image pre-processing methods in liver segmentation task
PublikacjaAutomatic liver segmentation of Computed Tomography (CT) images is becoming increasingly important. Although there are many publications in this field there is little explanation why certain pre-processing methods were utilised. This paper presents a comparison of the commonly used approach of Hounsfield Units (HU) windowing, histogram equalisation, and a combination of these methods to try to ascertain what are the differences...
-
Numerical investigations on early indicators of fracture in concrete at meso-scale.
PublikacjaFracture is a major reason of the global failure of concretes. The understanding of fracture is important to ensure the safety of structures and to optimize the material behaviour. In particular an early prediction possibility of fracture in concretes is of major importance. In this paper, concrete fracture under bending was numerically analysed using the Discrete Element Method (DEM). The real mesoscopic structure of a concrete...
-
Accelerating Video Frames Classification With Metric Based Scene Segmentation
PublikacjaThis paper addresses the problem of the efficient classification of images in a video stream in cases, where all of the video has to be labeled. Realizing the similarity of consecutive frames, we introduce a set of simple metrics to measure that similarity. To use these observations for decreasing the number of necessary classifications, we propose a scene segmentation algorithm. Performed experiments have evaluated the acquired...
-
GPU based implementation of Temperature-Vegetation Dryness Index for AVHRR3 Satellite Data
PublikacjaPaper presents an implementation of TVDI (Temperature-Vegetation-Dryness Index) algorithm on GPU (Graphics Processing Unit). Calculation of this index is based on LST (Land Surface Temperature) and NDVI (Normalized Difference Vegetation Index). Discussed results are based on multi-spectral imagery retrieved from AVHRR3 sensors for area of Poland. All phases of TVDI implementation on GPU are modified in respect to CUDA platform....
-
Image simulation and annotation for color blinded
PublikacjaIn this paper methods for image simulation as seen by a color blinded and a method for constructing images of perceived color difference are presented. The work is also focused on the interactive color description of an image contents. As a result, the individuals having problems with color discrimination can identify colors in an image.W artykule prezentowane są metody symulacji kolorów w obrazach postrzeganych przez osoby ze...
-
Multimodal English corpus for automatic speech recognition
PublikacjaA multimodal corpus developed for research of speech recognition based on audio-visual data is presented. Besides usual video and sound excerpts, the prepared database contains also thermovision images and depth maps. All streams were recorded simultaneously, therefore the corpus enables to examine the importance of the information provided by different modalities. Based on the recordings, it is also possible to develop a speech...
-
In vivo imaging of the human eye using a two-photon excited fluorescence scanning laser ophthalmoscope
PublikacjaBACKGROUND. Noninvasive assessment of metabolic processes that sustain regeneration of human retinal visual pigments (visual cycle) is essential to improve ophthalmic diagnostics and to accelerate development of new treatments to counter retinal diseases. Fluorescent vitamin A derivatives, which are the chemical intermediates of these processes, are highly sensitive to UV light; thus, safe analyses of these processes in humans...
-
Computer-Aided Detection of Hypertensive Retinopathy Using Depth-Wise Separable CNN
PublikacjaHypertensive retinopathy (HR) is a retinal disorder, linked to high blood pressure. The incidence of HR-eye illness is directly related to the severity and duration of hypertension. It is critical to identify and analyze HR at an early stage to avoid blindness. There are presently only a few computer-aided systems (CADx) designed to recognize HR. Instead, those systems concentrated on collecting features from many retinopathy-related...
-
INVESTIGATION OF BOTTOM SEDIMENT STRATIFICATION
PublikacjaThe main goal of this paper is to find a method to assess remotely the type of sediments on the basis of the signal received using the parametric echosounder. The images of the sea bottom sediments taken by a parametric echosounder show the fine structure of the upper layer of the bottom sediments. The depth of penetration depends on the type of sediment, in fact on the attenuation of acoustic wave. In the paper the method of distinguishing...
-
Configurational and Conformational Preferences in Stereoselective Acylations of N-Methyl-1,3-diaminopropane with Acyl Chlorides
PublikacjaEvidence for a stereoinduction profile of the reaction of N-methyl-1,3-diaminopropane with acyl chlorides was provided. A possibility to engage in intramol. CH2···HN and Cl···H-N interactions and the proton migration process to the methylamino group leads to the E secondary amides carrying the N···H+···N or N-H···N bridges, that show unusual spectroscopic images. Empirical relations between the ΔδC chem. shift differences,...
-
T1 RELAXATION TIME CALLIBRATION IN MAGNETIC RESONANCE IMAGING USING NANODIAMOND PHANTOMS
PublikacjaMRI is increasingly used in radiation treatment planning because of the excellent soft tissue contrast in the obtained images. It allows more precise definition of the boundaries between healthy tissues and those affected by cancerous lesions. To obtain good image quality, the difference in signal between two types of tissue must be significant despite the noise so it must be properly calibrated. This calibration includes...
-
Hierarchical 2-step neural-based LEGO bricks detection and labeling
PublikacjaLEGO 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...
-
Implementation of TVDI calculation for coastal zone
PublikacjaPaper will show an implementation of TVDI (Temperature-Vegetation-Dryness Index) algorithm on GPU (Graphics Processing Unit). Calculation of this index is based on LST (Land Surface Temperature) and NDVI (Normalized Difference Vegetation Index). Discussed results are based on multi-spectral imagery retrieved from AVHRR3 sensors for area of Poland, especially from region of Gdańsk coastal zone. All phases of TVDI implementation...
-
Smart Modeling of Maritime Vessels
PublikacjaCurrently, the market offers many visualization tools available to graphic designers, engineers, managers and academics working on maritime environments. The practice of visualization involves making and manipulating images that convey novel phenomena and ideas. Visual communication, together with virtual reality environments, is an emerging and rapidly evolving discipline. It brings great advantage over written word or voice alone,...
-
TOWARDS EXPLAINABLE CLASSIFIERS USING THE COUNTERFACTUAL APPROACH - GLOBAL EXPLANATIONS FOR DISCOVERING BIAS IN DATA
PublikacjaThe paper proposes summarized attribution-based post-hoc explanations for the detection and identification of bias in data. A global explanation is proposed, and a step-by-step framework on how to detect and test bias is introduced. Since removing unwanted bias is often a complicated and tremendous task, it is automatically inserted, instead. Then, the bias is evaluated with the proposed counterfactual approach. The obtained results...
-
An Overview of the Development of a Real-Time System for Endoscopic Video Classification
PublikacjaThe 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...
-
Longitudinal drug synergy assessment using convolutional neural network image-decoding of glioblastoma single-spheroid cultures
PublikacjaAbstract Background In recent years, drug combinations have become increasingly popular to improve therapeutic outcomes in various diseases, including difficult to cure cancers such as the brain cancer glioblastoma. Assessing the interaction between drugs over time is critical for predicting drug combination effectiveness and minimizing the risk of therapy resistance. However, as viability readouts of drug combination experiments...
-
Parameters optimization in medicine supporting image recognition algorithms
PublikacjaIn this paper, a procedure of automatic set up of image recognition algorithms' parameters is proposed, for the purpose of reducing the time needed for algorithms' development. The procedure is presented on two medicine supporting algorithms, performing bleeding detection in endoscopic images. Since the algorithms contain multiple parameters which must be specified, empirical testing is usually required to optimise the algorithm's...
-
Intelligent Autonomous Robot Supporting Small Pets in Domestic Environment
PublikacjaIn this contribution, we present preliminary results of the student project aimed at the development of an intelligent autonomous robot supporting small pets in a domestic environment. The main task of this robot is to protect a freely moving small pets against accidental stepping on them by home residents. For this purpose, we have developed the mobile robot which follows a pet and makes an alarm signal when a human is approaching....
-
Machine Learning Applied to Aspirated and Non-Aspirated Allophone Classification—An Approach Based on Audio "Fingerprinting"
PublikacjaThe purpose of this study is to involve both Convolutional Neural Networks and a typical learning algorithm in the allophone classification process. A list of words including aspirated and non-aspirated allophones pronounced by native and non-native English speakers is recorded and then edited and analyzed. Allophones extracted from English speakers’ recordings are presented in the form of two-dimensional spectrogram images and...
-
System for monitoring road slippery based on CCTV cameras and convolutional neural networks
PublikacjaThe slipperiness of the surface is essential for road safety. The growing number of CCTV cameras opens the possibility of using them to automatically detect the slippery surface and inform road users about it. This paper presents a system of developed intelligent road signs, including a detector based on convolutional neural networks (CNNs) and the transferlearning method employed to the processing of images acquired with video...
-
Water sorption and blistering of GFRP laminates with varying structures
PublikacjaThe microstructures, water absorption as well as blistering was studied for bi- and tri- axial glass fibre vinyl ester and polyester -matrix laminates coated with gel coat layer and uncoated. The effect of manufacturing technique on water sorption characteristics was considered. It was found that water sorption characteristics of GFRP depend on manufacturing methods. The least water intake was found for specimens having perfect...
-
Mesoscopic simulations of crack propagation in concrete using cohesive elements
PublikacjaThe paper presents results of two-and three-dimensional meso-scale simulations of fracture in notched concrete beams subjected to three-point bending test. Concrete was assumed as a 3-phase material composed of aggregate grains placed in the cement matrix with Interfacial Transitional Zones (ITZs) between then. In 2D simulations macro-voids were also taken into account. The particle distribution was taken from real concrete beams...
-
Sessile droplet contact angle of water–Al2O3, water–TiO2 and water–Cu nanofluids
PublikacjaThis work presents measurements of the contact angle of sessile droplets for three nanofluids, i.e. water–Al2O3, water–TiO2 and water–Cu. The plates made of glass, anodized aluminium and stainless steel of different roughness served as substrates. Ultrasonic vibration was used for 30–60 min in order to stabilise the dispersion of the nanoparticles. Nanoparticles were tested at the concentration of 0.01%, 0.1%, and 1% by weight....
-
Applications of Immersive 3D Visualization Lab
PublikacjaThe paper presents Immersive 3D Visualization Lab (at the Faculty of Electronics, Telecommunications and Informatics at Gdańsk University of Technology in Poland) and its applications prepared after its launch in December 2014. The main device of the lab is a virtual reality cubic cave with six screen-walls, where a person can navigate in virtual space using different controllers. The most sophisticated controller is a freely rotating...
-
Seafloor relief reconstruction from side scan sonar data
PublikacjaSide scan sonar is one of the most widely used imaging systems in the underwater environment. It is relatively cheap and easy to deploy, in comparison with more powerful sensors. Although side scan sonar does not provide seafloor bathymetry directly, its records are directly related to seafloor images. In the paper, the method for 3D seafloor relief reconstruction from side scan sonar data is presented. The method is based on the...
-
Reconstruction of 3D structure of positive corona streamer by local methods
PublikacjaThe computer algorithms were used for reconstruction of streamer 3D structure. We propose the 3D tree structure model of corona discharge streamer composed with nodes and edges between chosen couples of nodes, which enables easy computation of some important parameters ofstreamers. The 3D model can be derived directly from two projection images by global methods like evolutionary searching or particle simulations. In this paper...
-
Image Processing Techniques for Distributed Grid Applications
PublikacjaParallel approaches to 2D and 3D convolution processing of series of images have been presented. A distributed, practically oriented, 2D spatial convolution scheme has been elaborated and extended into the temporal domain. Complexity of the scheme has been determined and analysed with respect to coefficients in convolution kernels. Possibilities of parallelisation of the convolution operations have been analysed and the results...
-
Feature Reduction Using Similarity Measure in Object Detector Learning with Haar-like Features
PublikacjaThis paper presents two methods of training complexity reduction by additional selection of features to check in object detector training task by AdaBoost training algorithm. In the first method, the features with weak performance at first weak classifier building process are reduced based on a list of features sorted by minimum weighted error. In the second method the feature similarity measures are used to throw away that features...
-
Experimental investigations of damage evolution in concrete during bending by continuous micro-CT scanning
PublikacjaThe paper describes experimental investigation results of fracture in notched concrete beams under quasi-static three-point bending. To visualize 3D fracture in concrete under bending, an extended X-ray micro-computed tomography system was used, i.e. the tomography system SkyScan 1173 was connected to the loading machine ISTRON 5569. This combined system enabled to shot images of deforming concrete beams during a continuous deformation...
-
Stereo image visualization for a VISROBOT system
PublikacjaThe article describes a novel approach to robotic vision in mobile robot systems. The system implements a Visrobot system which implements a generic idea of using mobile robots for exploring an indoor environment. The task of such a robot is to visualize a stereo image properly for an operator. The system uses different stereo baseline values. Variable baseline can result in increasing depth resolution for distant objects. We assume...
-
Influence of Aggregating Window Size on Disparity Maps Obtained from Equal Baseline Multiple Camera Set (EBMCS)
PublikacjaThis paper is concerned with obtaining disparity maps on the basis of images from Equal Baseline Multiple Camera Set (EBMCS). EBMCS consists of a central camera and side cameras. Algorithms for obtaining disparity maps with the use of EBMCS take advantage of aggregating windows similarly to stereo matching algorithms for a stereo camera, a camera matrix or a camera array. The paper analyzes the influence of aggregating window size...
-
Intelligent system supporting diagnosis of malignant melanoma
PublikacjaMalignant melanomas are the most deadly type of skin cancers. Early diagnosis is a key for successful treatment and survival. The paper presents the system for supporting the process of diagnosis of skin lesions in order to detect a malignant melanoma. The paper describes the development process of an intel-ligent system purposed for the diagnosis of malignant melanoma. Presented sys-tem can be used as a decision support system...
-
Visual Object Tracking System Employing Fixed and PTZ Cameras
PublikacjaThe paper presents a video monitoring system utilizing fixed and PTZ cameras for tracking of moving objects. First type of camera provides image for background modelling, being employed for foreground objects localization. Estimated objects locations are then utilised for steering of PTZ cameras when observing targeted objects with high close-ups. Objects are classified into several classes, then basic event detection is being...