|
| |
|
|
Semantic description of the embedded device screen
Horák, Martin ; Richter, Miloslav (referee) ; Janáková, Ilona (advisor)
Tato diplomová práce se zabývá detekcí prvků uživatelského rozhraní na obrázku displejetiskárny za použití konvolučních neuronových sítí. V teoretické části je provedena rešeršesoučasně používaných architektur pro detekci objektů. V praktické čísti je probrána tvorbagalerie, učení a vyhodnocování vybraných modelů za použití Tensorflow ObjectDetectionAPI. Závěr práce pojednává o vhodnosti vycvičených modelů pro zadaný úkol.
|
|
|
Eye recognition system
Zvak, Martin ; Richter, Miloslav (referee) ; Petyovský, Petr (advisor)
This work deals with biometric methods, aiming at possibilities of eye recognition, and their application. It contains overview of biometrics evolution such as independent, fully accepted science during last decades and nowadays. Also it contains overview of most expanded biometric methods and takes a look on chosen method of eye recognition – iris recognition. Next aim of this work is design and description of implementation of suitable identification method for iris recognition, based on simple practices of image processing.
|
|
|
Image Motion Detection
Zítka, Michal ; Richter, Miloslav (referee) ; Horák, Karel (advisor)
This bachelor project dissert on detection of motion in image. Also sumary basic methods for detection of motion in image and method of segmentation using active contour. In practical section is attention paid to detection of motion in video camera screenshots.
|
|
|
System for stones counting
Krajíček, Pavel ; Richter, Miloslav (referee) ; Janáková, Ilona (advisor)
My bachelor thesis themed System for stones counting deals with a question of object recognition from a picture. This thesis enlarges on a suitable type of lighting during taking pictures of stones followed with methods of objects recognition. In the last phase an amount of stones occured in the picture is determined from recognized objects.
|
|
|
Static model of scene
Sikora, Jan ; Petyovský, Petr (referee) ; Richter, Miloslav (advisor)
This thesis deal with various methods of background detection and with it related motion detection in a scene. It's progressing from simplest methods to more comlex. For every one are reviewed the possibilities of using and her drawbacks. In introduction are described various types of scenes according to background and foreground type e.g . according to movement objects speed or presence of movement in background. Is proposed several common or specific improvements for obtaining better background even by using simple method. Next part of work solve real situation of shaking camera. There are tested two basic methods for optical stabilization. The first is registration of images by template matching. Alternative method used interest points (corners). Both methods are closely examinate and is sought best way to match following pictures. Except shaking of camera this work deal with rotating camera and in theory solve detection background from cameras placed on ridden car. Part of work is creation database of different types scenes
|
|
|
Detection of small object movements using camera
Udvardy, Bálint ; Janáková, Ilona (referee) ; Richter, Miloslav (advisor)
One of the basic problems in computer vision is motion detection and analysis in a given scene. This work focuses on detecting small changes in the image by using the moiré phenomenon. The main goal of this thesis is to detect different types of dislocations with algorithms used in computer vision. In this work synthetically created pictures are analysed, which were created with the mathematical model of a pinhole camera.
|
|
|
Tablet data acquisition
Zvak, Martin ; Babinec, Tomáš (referee) ; Richter, Miloslav (advisor)
Reliable authorization and authentication is now becoming necessary for many everyday activities such as boarding an aircraft, doing financial transactions etc. This work deals with the possibility of proposing an appropriate method of acquisition, processing and evaluation of data from the tablet itself, focusing on the possibility of obtaining data from the graphics tablet used in conjunction with a personal computer, for the purpose of verifying identity. It contains an overview of biometrics, as independent developing science over the last decades of the 20th century to the present, a summary of the development of a graphics tablet, the current situation, a brief overview of the most popular technologies used in their construction and distribution. Furthermore, it focuses on the design and implementation of appropriate procedures to obtain data provided by the Trust tablet TB - 7300, using available software development kit for the Microsoft Tablet PC, and a comparison of WinForms and WPF model (.NET Framework), with the declaration of suitability for the use of the second one. Last but not least, according to the response from the available set of users, this work focuses on the design of appropriate types of input data, test patterns, method and application for collecting and processing data from the graphic tablet, as well as their use for the verification of the identity of the individual. Ultimately, it is possible to discuss the accuracy of the method on a set of users, its suitability for use and shortcomings. Based on the analysis of the issue, the proposed method can be declared as under-optimal - it is safe, but inaccurate with low stability and repeatability. It is also important to accent that this is just the initial proposal, the use of applications in everyday life is for current phase impossible and the transition from laboratory conditions would require further development and improvement of the process.
|
|
|
Interpretation of Sign Language
Ficek, Dominik ; Honec, Peter (referee) ; Richter, Miloslav (advisor)
This work deals with image processing methods in persuit of sign language recognition and automatic interpretation in target sign language by artificial avatar. Following the theorethical introduction to the topic is a brief survey of current state of the art in sign language recognition. Proposed pipeline for the purposes of sign language recognition is based on the survey and latest state of the art hand and face landmark estimation methods. The designed pipeline is implemented and evaluated on suited metrics. As an application of this pipeline a system recognizing american sign langauge with automatic interpretation in czech sign language is built.
|
|
|
Universal microprocessor controller
Bartoň, Martin ; Richter, Miloslav (referee) ; Macho, Tomáš (advisor)
This work deals with the creation and construction of universal microprocessor controller on ARM7TDMI architecture. There are described various hardware features of this controller, design of the device, software equipment, used control algorithms and their implementation into the AT91SAM7X512 microcontroller. There is also described the verification of the controller on real plant.
|
guest ::
RSS feed.