|
|
Algorithms and Methods for Fusion of Matrix Sensors Data during Mapping Inside Building
Klečka, Jan ; Hlaváč, Václav (referee) ; Železný, Miloš (referee) ; Horák, Karel (advisor)
The dissertation is aimed at methods for simultaneous processing of various type sensor data into a virtual map. Conceptually the development was focused on algorithms for simultaneous localization and mapping. In theoretical part has the problem been addressed by the probabilistic approach. The practical part deals with a new approach called as partially collective mapping whose application leads to creation map in form of a set of simple geometrical entities which represents in piecewise manner a border between obstacles and free space.
|
|
|
Visual Object Tracking in Realtime
Kratochvíla, Lukáš ; Klečka, Jan (referee) ; Horák, Karel (advisor)
Sledování obecného objektu na zařízení s omezenými prostředky v reálném čase je obtížné. Mnoho algoritmů věnujících se této problematice již existuje. V této práci se s nimi seznámíme. Různé přístupy k této problematice jsou diskutovány včetně hlubokého učení. Představeny jsou reprezentace objektu, datasety i metriky pro vyhodnocování. Mnoho sledovacích algorimů je představeno, osm z nich je implementováno a vyhodnoceno na VOT datasetu.
|
|
|
Basics of Pedestrians Detection in Image by Machine Learning
Lučanský, Peter ; Klečka, Jan (referee) ; Horák, Karel (advisor)
Táto Bakalárska práce sa zaoberá významnou problematikou v oblasti počítačového videnia, ktorou je detekcia osôb/chodcov v obraze, za pomoci metod strojového učenia, spolu s jej možným využitím, vývojom a vysvetlením princípov. Taktiež sa zaoberá testovaním dnes najlepšieho dostupného algoritmu, pričom sa porovnávajú faktory ktoré vplívajú na kvalitu jeho činnosti. Na začiatku je problematika stručne popísaná, potom sa prejde k podrobným popisom dosiahnutých pokrokov. V nasledujúcej časti sú popísané dostupné datasety, ktoré by sa dali použiť pri tréningu detekčného algoritmu. V poslednom rade sú vykonané trénovacie procesy za rozličných podmienok, pričom sú jednotlivé výsledky porovnávané.
|
|
|
3D model
Krátký, Martin ; Klečka, Jan (referee) ; Richter, Miloslav (advisor)
The aim of this bachelor thesis is to create a spatial model from images capturing an object illuminated by a laser pattern. The thesis presents possible solutions to the problem using methods of active triangulation and stereoscopy. Methods are designed for laboratory and outdoor measurements. The requirements for the measuring workplace are defined and then the implemented workplace is presented. Each of the designed methods is tested by scanning a series of objects.
|
|
|
Pedestrians Detection in Traffic Environment by Machine Learning
Tilgner, Martin ; Klečka, Jan (referee) ; Horák, Karel (advisor)
Tato práce se zabývá detekcí chodců pomocí konvolučních neuronových sítí z pohledu autonomního vozidla. A to zejména jejich otestováním ve smyslu nalezení vhodné praxe tvorby datasetu pro machine learning modely. V práci bylo natrénováno celkem deset machine learning modelů meta architektur Faster R-CNN s ResNet 101 jako feature extraktorem a SSDLite s feature extraktorem MobileNet_v2. Tyto modely byly natrénovány na datasetech o různých velikostech. Nejlépší výsledky byly dosaženy na datasetu o velikosti 5000 snímků. Kromě těchto modelů byl vytvořen nový dataset zaměřující se na chodce v noci. Dále byla vytvořena knihovna Python funkcí pro práci s datasety a automatickou tvorbu datasetu.
|
|
|
3D reconstruction from multiple views
Sládeček, Martin ; Klečka, Jan (referee) ; Richter, Miloslav (advisor)
This thesis deals with the task of three-dimensional scene reconstruction using image data obtained from multiple views. It is assumed that intrinsic parameters of the utilized cameras are known. The theoretical chapters describe the basic priciples of individual reconstruction steps. Variuous possible implementaions of data model suitable for this task are also described. The practical part also includes a comparison of false keypoint correspondence filtering, implementation of polar stereo rectification and comparison of disparity map calculation methods that are bundled with the OpenCV library. In the final portion of the thesis, examples of reconstructed 3D models are presented and discussed.
|
|
|
Authentication with RFID tags
Klečka, Jan ; Dzurenda, Petr (referee) ; Malina, Lukáš (advisor)
The bachaleror thesis deals with issue of RFID elements and authentication systems used by elements of RFID and chip cards. The basic differences between active and passive RIFD tags are decsribed. The thesis describes basic attacks which can threat communication between reader and RFID device in wireless comunication. In the thessis is included measuring of passive RFID tags. Results of the measurement are presented by tables and graphs. Main goal of this thesis aims at the implementation of the authentication protocol which is mainly for processor cards. Because of the processor we can use cryptography where is neccessary computing power. We can ensure secure data transfer and authentication. The thesis describes 2 schemes of one authentication protocol, which were modificated for BasicCard Enviroment. In the first scheme client(card) authenticate to the terminal using secure AES 256 bit data transfer and verifier is the terminal side. Data which needs to be exchanged is secured by AES 256 bit. In the second scheme the card side is a verifier and the card generates such data to verify that the other party is able to authenticate.
|
|
|
Comparison Of Different Approaches To
Adaptation Of Unscented Kalman Filter
For Non-Odometry Slam
Klečka, Jan
This paper is aimed at utilizing of an Unscented Kalman filter (UKF) for application on non-odometry SLAM problem. Non-odometry SLAM problem lacks state transitions function for part of state vector so theoretical part of the paper describes the derivation of three proposed modifications. The experimental part then shows experiments which evaluate modifications performance in reference to Extended Kalman filter. Experiments are based on simulation of monocular photogrammetry reconstruction problem.
|
|
|
Continuous Realization of Monocular Structure from Motion
Klečka, Jan
This paper describes a concept and practical aspects of a realization of an algorithm used for 3D scene reconstruction from multiple views captured by single moving camera. Position and orientation of the camera are a priory unknown and are calculated on-the-fly from tracked point correspondences.
|
|
|
High data rate image processing using CUDA/OpenCL
Sedláček, Filip ; Klečka, Jan (referee) ; Honec, Peter (advisor)
The main objective of this research is to propose optimization of the defect detection algorithm in the production of nonwoven textile. The algorithm was developed by CAMEA spol. s.r.o. As a consequence of upgrading the current camera system to a more powerful one, it will be necessary to optimize the current algorithm and choose the hardware with the appropriate architecture on which the calculations will be performed. This work will describe a usefull programming techniques of CUDA software architecture and OpenCL framework in details. Using these tools, we proposed to implement a parallel equivalent of the current algorithm, describe various optimization methods, and we designed a GUI to test these methods.
|
guest ::
RSS feed.