|
|
Conversion of Raster-Curve into Vector Representation
Král, Jiří ; Sumec, Stanislav (referee) ; Beran, Vítězslav (advisor)
In my process of tracing I deal with converting an input grayscale image into a vector one trying to keep as big similarity with the input image as possible. Tracing is carried out with the help of curve approximation even if the approximation is possible only with line elements, that is to say the curve in raster. Therefore it is necessary to extract the line elements from the input image. We can do it in two different ways according to two different objects in the image. The first group is represented by thin, ablong objects which are substituted by their skeleton. The second group is represented by large objects which are susbstituted by their contour. The found lines are then divided into such parts which can be easily curve approximated. Resulting curves are then only depicted into the output by suitable raster method.
|
|
|
Wireless Sensor Network with Arduino Components
Výborný, Filip ; Král, Jiří (referee) ; Samek, Jan (advisor)
The goal of this bachelor's thesis is to introduce the possibility of using Arduino hardware platform as an aid to the creation of wireless sensor networks, focusing especially on the routing of measured data issues from sensor nodes to the base station. My thesis deals with the design and implementation of wireless sensor networks from suitable Arduino components, based on the XBee wireless communication modules which are introduced in the second chapter. The third chapter discusses the selection of a suitable communication protocol, the relatively unknown protocol DigiMesh. The final chapter covers the design of the actual wireless network which has been tested.
|
|
|
BOIDS Method for Swarm Simulation
Burda, Radek ; Král, Jiří (referee) ; Zbořil, František (advisor)
This work primarily deals with C.Reynolds's model of flocking -BOIDS - and uses the model as a basis for creating a swarm simulation. It discusses methods for obstacle avoidance and principle of forces arbitration (flocking rules, obstacle avoidance and goal satisfying) to properly avoid conflicting of behaviours. Furhermore some of other approaches to flocking simulation are mentioned while their pros and cons are taken up. Last but not least proceeding of creating a graphic environment in Blender for demonstrating boids behavior in the final 3D application is described.
|
|
| |
|
|
Simulation of Cooperation of Agents in Jason Environment
Kříž, Jakub ; Zbořil, František (referee) ; Král, Jiří (advisor)
This work deals with creation of simulator of multi-agent system in which agents cooperate. Reader is introduced to basics of agent systems, their creating and modeling. Design of environment in which agents exist and task they solve is described in this work. There are designed and implemented three levels of racional BDI inteligence of agents with different level of cooperation - without cooperation, weak cooperation and complex cooperation. Whole system is implemented in Jason framework and extensively tested. Behavior of agents is analyzed and levels of intelligence are compared based on simulation experiments. Achieved results prove dominance of team in which agents cooperate more.
|
|
| |
|
|
Machine Learning in Image Classification
Král, Jiří ; Španěl, Michal (referee) ; Hradiš, Michal (advisor)
This project deals vith analysis and testing of algorithms and statistical models, that could potentionaly improve resuts of FIT BUT in ImageNet Large Scale Visual Recognition Challenge and TRECVID. Multinomial model was tested. Phonotactic Intersession Variation Compensation (PIVCO) model was used for reducing random e ffects in image representation and for dimensionality reduction. PIVCO - dimensionality reduction achieved the best mean average precision while reducing to one-twenyth of original dimension. KPCA model was tested to approximate Kernel SVM. All statistical models were tested on Pascal VOC 2007 dataset.
|
|
|
Facial Expression Recognition
Král, Jiří ; Kršek, Přemysl (referee) ; Španěl, Michal (advisor)
Many views to facial expression recognition exist. This work presents one of approaches. Existing methods of human face representation by model are discussed. The AAM method, where final appearance model is created from model of shape and model of texture is proposed. Model of shape and model of texture is created by statistic analysis. Using this representation, an effective method is achieved that is complexity of information for searched face in static image. Choice and combination of suitable features for classification of facial expression is principle for facial expression recognition based on AAM. Two approaches of facial expression classification are compared. Classification based on LDA and classification based on SVM. These methods with necessary face localization using AdaBoost form an automated face recognizer in image.
|
|
|
Social Network Analysis and Simulations
Vorlová, Pavla ; Král, Jiří (referee) ; Samek, Jan (advisor)
This diploma thesis is focusing on description of processing social network analysis, design and implementation of a model that simulates a particular social network and its analysis. Social networks are modern and very used in this time. They are very good point for exploring. This project deal with static analysis social network, where social network is constructed by graph. We nd out di erent properties of single component and than we establish signi cance of them. Relationships between components are important too for us, because they have a big influence on propagation information in network. Structural properties figure out existence of di fferent communities. We simulate social network with multi-agent systems, they are desirable for represent changes in network. Multi-agent systems have implemented a simulation model that represents a particular social network. His behaviour was analyzed and examinated by chosen methods.
|
|
|
Coevolution of Cartesian Genetic Algorithms and Neural Networks
Kolář, Adam ; Král, Jiří (referee) ; Zbořil, František (advisor)
The aim of the thesis is to verify synergy of genetic programming and neural networks. Solution is provided by set of experiments with implemented library built upon benchmark tasks. I've done experiments with directly and also indirectly encoded neural netwrok. I focused on finding robust solutions and the best calculation of configurations, overfitting detection and advanced stimulations of solution with fitness function. Generally better solutions were found using lower values of parameters n_c and n_r. These solutions tended less to be overfitted. I was able to evolve neurocontroller eliminating oscilations in pole balancing problem. In cancer detection problem, precision of provided solution was over 98%, which overcame compared techniques. I succeeded also in designing of maze model, where agent was able to perform multistep tasks.
|
guest ::
RSS feed.