Yes (208)
No (111)
Used filters
(3)
Number of found documents:
319
Published from
to
SUPER-RESOLUTION MICROSCOPY LIVE CELL IMAGING AND IMAGE ANALYSIS
Klíma, Miloš; Fliegel, Karel; Lasser, Theo; Lukeš, Tomáš
2017 - English
Novel fundamental research results provided new techniques going beyond the diffraction
limit. These recent advances known as super-resolution microscopy have been awarded by the
Nobel Prize as they promise new discoveries in biology and live sciences. All these techniques
rely on complex signal and image processing. The applicability in biology, and particularly for
live cell imaging, remains challenging and needs further investigation.
Focusing on image processing and analysis, the thesis is devoted to a significant enhancement
of structured illumination microscopy (SIM) and super-resolution optical fluctuation
imaging (SOFI)methods towards fast live cell and quantitative imaging. The thesis presents
a novel image reconstruction method for both 2D and 3D SIM data, compatible with weak
signals, and robust towards unwanted image artifacts. This image reconstruction is efficient
under low light conditions, reduces phototoxicity and facilitates live cell observations. We
demonstrate the performance of our new method by imaging long super-resolution video
sequences of live U2-OS cells and improving cell particle tracking. We develop an adapted
3D deconvolution algorithm for SOFI, which suppresses noise and makes 3D SOFI live cell
imaging feasible due to reduction of the number of required input images. We introduce a
novel linearization procedure for SOFI maximizing the resolution gain and show that SOFI
and PALMcan both be applied on the same dataset revealing more insights about the sample.
This PALMand SOFI concept provides an enlarged quantitative imaging framework, allowing
unprecedented functional exploration of the sample through the estimation of molecular
parameters. For quantifying the outcome of our super-resolutionmethods, the thesis presents
a novel methodology for objective image quality assessment measuring spatial resolution
and signal to noise ratio in real samples. We demonstrate our enhanced SOFI framework
by high throughput 3D imaging of live HeLa cells acquiring the whole super-resolution 3D
image in 0.95 s, by investigating focal adhesions in live MEF cells, by fast optical readout
of fluorescently labelled DNA strands and by unraveling the nanoscale organization of CD4
proteins on a plasma membrane of T-cells. Within the thesis, unique open-source software
packages SIMToolbox and SOFI simulation tool were developed to facilitate implementation
of super-resolution microscopy methods.
Available in digital repository of ČVUT.
SUPER-RESOLUTION MICROSCOPY LIVE CELL IMAGING AND IMAGE ANALYSIS
Novel fundamental research results provided new techniques going beyond the diffraction limit. These recent advances known as super-resolution microscopy have been awarded by the Nobel Prize as they ...
Example-based Rendering of Hand-drawn Content
Sýkora, Daniel; Fišer, Jakub
2017 - English
In this dissertation, we focus on the capabilities of example-based synthesis with a
particular interest in hand-drawn appearance. We propose several approaches for creating
new digital imagery from examples without making any domain-speci c assumptions
and being media-agnostic. New methods for content-creation and non-photorealistic stylization
are presented together with examples of their practical usage in contemporary
digital production pipelines.
Available in digital repository of ČVUT.
Example-based Rendering of Hand-drawn Content
In this dissertation, we focus on the capabilities of example-based synthesis with a particular interest in hand-drawn appearance. We propose several approaches for creating new digital imagery from ...
Local navigation methods for metal detectors
Ripka, Pavel; Roháč, Jan; Nováček, Petr
2017 - English
This dissertation addresses the implementation
of navigation capabilities for
metal detection, in particular landmines.
The precise aim of this thesis is to find
a professional handheld metal detector
search head position. This allows the creation
of a magnetic image of the area
searched, which opens a new prospective
to increase the discrimination ability of
metal detection between conductive scrap
and the surveillance of unexploded ordnance.
This thesis has two parts. The first and
main part focuses on the enhancement of
professional handheld metal detector discrimination
functionality. This was the
main stream of my research work. The
second part deals with position estimation
as a tool to reach the objectives of the first
part. The navigation is based on low-cost
MEMS inertial sensors of angular rate and
acceleration, which have minimum metal
content. The required precision, in centimetres,
can only be achieved by using
calibration and signal enhancement procedures.
The utilization of the magnetic
position markers directly detected by a
metal detector itself is my main original
scientific contribution to the knowledge
in the scientific area of aiding navigation.
Position markers with specific characteristics
are placed in a known location
and are recognised by a metal detector
during its overhead movement.
The metal position markers described
are a connection link between the navigation
part and the part dealing with metal
detector enhancement. Their usage is only
enabled due to previous research made on
the signal dependency of the metal detector’s
output on a detected object’s material,
size and distance from the detector’s
search head. Metal position markers are
a by-product of this research, which focuses
in particularly on the improvement
of metal detector discrimination abilities.
The ability of metal detector discrimination
can also be increased by using
several excitation signal frequencies. The
methods published up to now use multitone
signals composed of sinewaves with
two to four frequencies. In the presented
papers we show that using sin(x)/x excitation
signal brings new possibilities in
this area.Tato dizertacní práce se zabývá implementací
navigacních schopností na detektory
kovových predmetu a zejména min.
Urcení pozice detekcní hlavy profesionálního
detektoru kovu je presný cíl této dizertacní
práce. Známá pozice umožní vytvorení
magnetického obrazu zkoumané
oblasti a tím zvýší diskriminacní schopnost
detektoru kovu odlišit nevybuchlou
munici od vodivého odpadu.
Dizertacní práce má dve cástí. První,
hlavní cást, je zamerena na zvýšení diskriminacních
schopností profesionálního
rucního detektoru kovu. Toto byl hlavní
smer mé výzkumné cinnosti. Druhá cást
se zabývá odhadem pozice detektoru jako
nástroje k dosažení cíle první cásti práce.
Navigacní schopnost je založena na využití
levných inerciálních senzoru typu
MEMS merících zrychlení a úhlovou rychlost,
u kterých byl kladen duraz na minimální
obsah kovu. Vyžadované presnosti
jednoho centimetru je možné dosáhnout
pouze za pomoci kalibrace a zpresnujících
postupu. Využití magnetických pozicních
znacek prímo detekovaných detektorem
kovu je mým vlastním vedeckým prínosem
vedecké znalosti v oblasti navigacního
zpresnování.
Popisované kovové pozicní znacky jsou
pojítkem mezi navigacní cástí a cástí zabývající
se vylepšením detektoru kovu. Jejich
využití je umožneno pouze díky predchozímu
výzkumu provádeného na závislosti
výstupního signálu detektoru kovu na materiálu
detekovaného objektu, jeho velikosti
a vzdálenosti od detekcní hlavy detektoru
kovu. Kovové pozicní znacky jsou
vedlejším produktem tohoto výzkumu,
který cílil predevším na zvýšení diskriminacních
schopností detektoru kovu.
Zvýšení diskriminacních schopností detektoru
kovu muže být dále docíleno i za
pomoci využití více frekvencí použitých v
budícím signálu. Publikované metody využívají
více tónového signálu vytvoreného
ze sinusoid o dvou až ctyrech frekvencích.
V prezentovaných výsledcích jsme ukázali,
že využití sin (x)/x budicího signálu prináší
nové možnosti v této oblasti.
Available in digital repository of ČVUT.
Local navigation methods for metal detectors
This dissertation addresses the implementation of navigation capabilities for metal detection, in particular landmines. The precise aim of this thesis is to find a professional handheld metal ...
Short-Term Visual Object Tracking in Real-Time
Matas, Jiří; Vojíř, Tomáš
2017 - English
In the thesis, we propose two novel short-term object tracking methods, the Flock of
Trackers (FoT) and the Scale-Adaptive Mean-Shift (ASMS), a framework for fusion of
multiple trackers and detector and contributions to the problem of tracker evaluation
within the Visual Object Tracking (VOT) initiative.
The Flock of Trackers partitions the object of interest to an equally sized parts. For
each part, the FoT computes an optical flow correspondence and estimates its reliability.
Reliable correspondences are used to robustly estimates a target pose using RANSAC
technique, which allows for range of complex rigid transformation (e.g. affine transformation)
of a target. The scale-adaptive mean-shift tracker is a gradient optimization
method that iteratively moves a search window to the position which minimizes a distance
of a appearance model extracted from the search window to the target model. The
ASMS propose a theoretically justified modification of the mean-shift framework that
addresses one of the drawbacks of the mean-shift trackers which is the fixed size search
window, i.e. target scale. Moreover, the ASMS introduce a technique that incorporates
a background information into the gradient optimization to reduce tracker failures in
presence of background clutter.
To take advantage of strengths of the previous methods, we introduce a novel tracking
framework HMMTxD that fuses multiple tracking methods together with a proposed
feature-based online detector. The framework utilizes a hidden Markov model
(HMM) to learn online how well each tracking method performs using sparsely ”annotated”
data provided by a detector, which are assumed to be correct, and confidence
provided by the trackers. The HMM estimates the probability that a tracker is correct in
the current frame given the previously learned HMM model and the current tracker confidence.
This tracker fusion alleviates the drawbacks of the individual tracking methods
since the HMMTxD learns which trackers are performing well and switch off the rest.
All of the proposed trackers were extensively evaluated on several benchmarks and
publicly available tracking sequences and achieve excellent results in various evaluation
criteria. The FoT achieved state-of-the-art performance in the VOT2013 benchmark,
finishing second. Today, the FoT is used as a building block in complex applications
such as multi-object tracking frameworks. The ASMS achieved state-of-the-art results
in the VOT2015 benchmark and was chosen as the best performing method in terms
of a trade-off between performance and running time. The HMMTxD demonstrated
state-of-the-art performance in multiple benchmarks (VOT2014, VOT2015 and OTB).
The thesis also contributes, and provides an overview, to the Visual Object Tracking
(VOT) evaluation methodology. This methodology provides a means for unbiased
comparison of different tracking methods across publication, which is crucial for advancement
of the state-of-the-art over a longer timespan and also provides a tools for
deeper performance analysis of tracking methods. Furthermore, a annual workshops
are organized on major computer vision conferences, where the authors are encouraged
to submit their novel methods to compete against each other and where the advances in
the visual object tracking are discussed.
Available in digital repository of ČVUT.
Short-Term Visual Object Tracking in Real-Time
In the thesis, we propose two novel short-term object tracking methods, the Flock of Trackers (FoT) and the Scale-Adaptive Mean-Shift (ASMS), a framework for fusion of multiple trackers and detector ...
Multi-view Facial Landmark Detection
Franc, Vojtěch; Uřičář, Michal
2017 - English
In this thesis, we tackle the problem of designing a multi-view facial landmark detector
which is robust and works in real-time on low-end hardware. Our landmark detector
is an instance of the structured output classi ers describing the face by a mixture
of tree based Deformable Part Models (DPM). We propose to learn parameters of
the detector by the Structured Output Support Vector Machine algorithm which, in
contrast to existing methods, directly optimizes a loss function closely related to the
standard evaluation metrics used in landmark detection. We also propose a novel
two-stage approach to learn the multi-view landmark detectors, which provides better
localization accuracy and signi cantly reduces the overall learning time. We propose
several speedups that enable to use the globally optimal prediction strategy based on
the dynamic programming in real time even for dense landmark sets. The empirical
evaluation shows that the proposed detector is competitive with the current state-ofthe-
art both regarding the accuracy and speed.
We also propose two improvements of the Bundle Method for Regularized Risk Minimization
(BMRM) algorithm which is among the most popular batch solvers used
in structured output learning. First, we propose to augment the objective function
by a quadratic prox-center whose strength is controlled by a novel adaptive strategy
preventing zig-zag behavior in the cases when the genuine regularization term is weak.
Second, we propose to speed up convergence by using multiple cutting plane models
which better approximate the objective function with minimal increase in the computational
cost. Experimental evaluation shows that the new BMRM algorithm which
uses both improvements speeds up learning up to an order of magnitude on standard
computer vision benchmarks, and 3 to 4 times when applied to the learning of the
DPM based landmark detector.
v
Available in digital repository of ČVUT.
Multi-view Facial Landmark Detection
In this thesis, we tackle the problem of designing a multi-view facial landmark detector which is robust and works in real-time on low-end hardware. Our landmark detector is an instance of the ...
Enhancement of the selected management and planning quality control tools
Mach, Pavel; Žák, Vratislav; Povolotskaya, Evgenia
2017 - English
A detailed analysis of three management and planning quality control tools – affinity diagram, arrow diagram and process decision program chart – was performed. Then fuzzy logic was used for increase the yield of information of these tools. All three tools shown up to be flexible for fuzzy approach application.
As the first, an application of an arrow diagram with a support of fuzzy logic to solution of mobile phone functionality problem at a service center for repair of mobile phones is presented. For the smooth course of negotiation between a client and the mobile phone reparation service there is a need to estimate the time of the problem solution as precisely as possible. Fuzzy logic as a supporter for an arrow diagram was used in order to determine the time interval of mobile reparation process more accurately in comparison with a standard approach. Better estimation of the time of a step “Diagnostics with the goal to find a cause of a problem” using the fuzzy logic is demonstrated.
Then, an affinity diagram formed for a process of adhesive assembly of electronic components on the printed circuit board was constructed. Fuzzy approach showed the degree of an influence of the failures in some operations of the process to the failures in the following operations of this process. Non-obvious failures that could be met during a testing step were estimated and their significance was calculated.
Process decision program chart (PDPC) created for assembly of electronic components on a printed circuit board based on wave soldering was also presented. The process decision program chart was constructed to identify risks and develop counter-measures (preventive actions). Volume of information, which should be involved into such the diagram, is very high. Fuzzy approach was used to find information of lower importance, which can occur, and thereby reduce the amount of entering information for processing of the PDPC diagram.
The use of fuzzy logic is, with respect to the range of the thesis, documented on selected steps used in the processing of selected managerial quality control tools. The approach is generic and can be used for any technological or other type of an operation or a process to obtain deeper insight into the problem and better estimation of the course of an event under analysis.
Available in digital repository of ČVUT.
Enhancement of the selected management and planning quality control tools
A detailed analysis of three management and planning quality control tools – affinity diagram, arrow diagram and process decision program chart – was performed. Then fuzzy logic was used for increase ...
Identification of femtocells in mobile networks
Bešťák, Robert; Sedláček, Marek
2017 - English
The evolving mobile networks are requested to convey increasing data traffic as popularity of online services together with affordability of mobile devices is growing. One solution to mobile carriers, which can help them quickly deploy small base stations (BS) ensuring great indoor coverage with minimum costs, and high data rate capability, is femtocell technology. However, standard deployment techniques are unsatisfactory for these type of BSs. There are two main reasons for that. Firstly, femtocells will be deployed in great numbers. Secondly, they are deployed by users and are portable. It means their position is not known in advance, and can vary in time. Therefore, femtocells have to implement self-configuration principles. Physical Cell Identity is one of the most important parameters to be chosen automatically under defined conditions. It is crucial parameter, which allows them to convey a communication between a user equipment and a core network. A study on Physical Cell Identity issues in mobile networks with femtocells is presented in my thesis. For this purpose, I created two different models of femtocells deployment and deal with a collision and a confusion. They are two main problems, which threaten proper Physical Cell Identity assignment in mobile networks. Outputs of the thesis serves for better understanding of interrelations between differently placed femtocells in term of collision and confusion issue and as the basis to design the framework handling Physical Cell Identity allocation. The simulations conducted on proposed models were utilized to obtain probability characteristics and indicators based on graph theory. In the evaluation section, I appoint several characteristics as probability of collision, probability of confusion and maximal number of neighbourhood cells and some others to support solution of collision and confusion issue. I use results of evaluation and layout the framework for automated Physical Cell Identity assignment with two different approaches, the distributed one, and the centralized one. Since, femtocells are subcategory of small cells so findings, mentioned in this thesis, can also be used for other types of small cells.
Available in digital repository of ČVUT.
Identification of femtocells in mobile networks
The evolving mobile networks are requested to convey increasing data traffic as popularity of online services together with affordability of mobile devices is growing. One solution to mobile carriers, ...
Algorithms for Complex Bipedal Walking
Čelikovský, Sergej; Dolinský, Kamil
2017 - English
This thesis studies the problem of parameter estimation, model identi-
fication and state estimation for underactuated bipedal walking robots.
Two main results were developed. The first result is a novel identification
method suited for this problem. The second result is the extension of
existing algorithms for state estimation to the case of the hybrid model
of an underactuated walking robot.
The identification method takes advantage of the linear structure of the
model with respect to estimated parameters. The resulting estimator
is calculated iteratively and maximizes the likelihood of the data. The
method was tested on both simulated and experimental data. Simulations
were carried out for an underactuated walking robot with a distance meter
to measure absolute orientation. Laboratory experiments were carried
out on a leg of a laboratory walking robot model equipped with a threeaxis
accelerometer and gyroscope to measure absolute orientation. The
method performs favorably in comparison with other benchmark estimation
algorithms and both the simulations and the laboratory experiments
con rmed its high potential for the use in identi cation of underactuated
robotic walkers.
The state estimators were applied to estimate the absolute orientation
of an underactuated walking robot in the presence of impacts which occur
when the leg of the robot hits the ground. The proposed estimation
scheme was tested on simulations of a 3-link robot and shows that proposed
extensions yields improved estimation performance.
Keywords:
walking robots; maximum likelihood estimation; identification; state estimation
Available in digital repository of ČVUT.
Algorithms for Complex Bipedal Walking
This thesis studies the problem of parameter estimation, model identi- fication and state estimation for underactuated bipedal walking robots. Two main results were developed. The first result is a ...
Measurement of fast-flying targets in near-zone
Hudec, Přemysl; Jeník, Vojtěch
2017 - English
This thesis deals with the problem of detection, identification and tracking of fast flying objects
close to radar sensors, which reflects a current set-up in radar based active defense systems
(ADS). Furthermore, a comprehensive noise analysis of CW radar is presented.
Various configurations of CW radar are proposed and evaluated – from multiple installations of
identical CW radar sensors to a set of receivers sharing the same HW component (e.g. local
oscillator or the whole transmitter) – with the aim of finding the best and simplest possible
configuration to fulfill the demanding requirements of ADS. Since the radar sensors are very
complicated systems, the best solution cannot be identified without considering individual
requirements on the system and available technology (i.e. frequency range, modulation, etc.).
Therefore the thesis only concludes the principal properties of the proposed configurations.
Another topic investigates available methods for in-field radar sensor testing and calibration.
With regard to various methods and their limitations, the selected solution uses an antenna with
an electronically controlled phase shifter and short-end stub.
The last part is dedicated to the comprehensive noise analysis of a homodyne CW radar system
with quadrature demodulation, which includes the amplitude and phase noise of local oscillator,
noise of linear and quasi-linear RF components and noise of low frequency circuits. Even though
the analysis has been developed on homodyne CW radar, the results are applicable to a much
wider range of CW radar sensors. The follow-up research into this topic may be focused on
behavior of SDR-based radar systems.
Available in digital repository of ČVUT.
Measurement of fast-flying targets in near-zone
This thesis deals with the problem of detection, identification and tracking of fast flying objects close to radar sensors, which reflects a current set-up in radar based active defense ...
Reliability of optical structures in harsch environments
Zvánovec, Stanislav; Komanec, Matěj; Bohata, Jan
2017 - English
Networks of decades-old optical fibers created for operating optical communications are still in use today and being applied to the harsh environments of avionics and military, marine, and space systems. Consequently, an analysis of optical component reliability and a description of the aging process is called to predict the behavior of these fibers in specific areas of utilization. In this thesis, harsh environments for optical components are investigated and a methodology for the long-term monitoring of optical fiber parameters for different service conditions is presented. Furthermore, several optical systems are tested over aged optical infrastructure.
Keywords:
optical fiber; harsh environment; aging; radio over fiber; free-space optics
Available in digital repository of ČVUT.
Reliability of optical structures in harsch environments
Networks of decades-old optical fibers created for operating optical communications are still in use today and being applied to the harsh environments of avionics and military, marine, and space ...
NRGL provides central access to information on grey literature produced in the Czech Republic in the fields of science, research and education. You can find more information about grey literature and NRGL at service web
Send your suggestions and comments to nusl@techlib.cz
Provider
Other bases