National Repository of Grey Literature 22 records found  1 - 10nextend  jump to record: Search took 0.00 seconds. 
Automatic Retinal Image Quality Analysis
Dovbush, Andrii ; Rydlo, Štěpán (referee) ; Kavetskyi, Andrii (advisor)
The work was carried out in the context of the development of an autonomous ophthalmological examination centre. In addition to the accuracy of assessing image quality, a significant factor was the speed of calculation of key image quality parameters, which would allow real-time operation of the centre. Based on recent researches and scientific discoveries, there was a need of additional study on algorithms' time performance. The main goal of the work was to create an algorithm for assessing image quality, based on the best combination of image quality parameters in the context of effectiveness-to-time performance. This eventually led to the development of a tool for analysing retinal image quality features and the subsequent analysis of selected combinations of IQA algorithms and image channels. Two methods for selecting the best combinations of algorithms and image channels were chosen: the absolute performance method and the relative performance method. As a result, their weaknesses were found and a possible solution to these problems was proposed by using the best-coverage approach. Approximately 150 possible combinations of quality feature and channel extraction algorithms have been collected and analysed. Among the selected algorithms it was also possible to determine the algorithms dependent on the level of image noise and also the influence of resolution downscaling on the overall performance of the different algorithms for quality features. The effect of noise on the Sobel edge detector and the Canny edge detector was explained.
Reconstruction and Enhancement of Damaged Parts of Fingerprint Images
Špila, Andrej ; Rydlo, Štěpán (referee) ; Heidari, Mona (advisor)
Táto práca sa zaoberá problémom rekonštrukcie snímkov odtlačkov prsta so zameraním na neobnovyteľné oblasti poškodené rôznymi kožnými ochoreniami. Generatívne súperi- ace siete s trénovateľnou konvolučnou vrstvou s gaborovými filtrami bola natrénovaná na dátovej sade reálnych snímkov odtlačkov prsta. Práca predvádza že natrénovaný model vie spoľahlivo rekonštruovať malé oblasti ľubovoľného tvaru a v prípade väčších oblastí, globálne skóre kvality rekonštruovaného odtlačku prsta získané využitím softvéru NIST biometric image software sa v porovnaní s originálnym snímkom navýšilo. Je navrhnutý štandardizovaný formát pre snímky odtlačkov prsta ktorý pomohol stabilizovat trénovanie generatívnych súperiacich sietí.
Animal Identification Based on Biometric Information
Plička, Maxim ; Šátek, Václav (referee) ; Rydlo, Štěpán (advisor)
This bachelor thesis is focused on obtaining information from a photograph of animal's muzzle pattern that will be used to identify the individual. The main goal is to propose and implement an application that can extract this information from the input photo and then output it, or find the photo in the database that most closely matches the input photo. The main purpose of the testing was to determine the accuracy of extracting this information. The proposed approach achieved 82.71% accuracy with a database size of 171 photos. The photographs belonged to deer and does.
Blood Vessels Detection in the Fingers Area for Liveness Detection
Šulavík, Jan ; Goldmann, Tomáš (referee) ; Rydlo, Štěpán (advisor)
Biometrics as a part of computer technology are recently rising in popularity and various biometric methods are becoming a part of everyday human life. This work in the first place informs the reader on the basics of biometrics in computer technology and further continues to discuss the use of veins in this area as veins are for their uniqueness considered as an ideal human physiological trait for use in biometrics. It is described in what manner is this biometric characteristic being captured. Further more the work describes some image processing methods for extracting biometric data from the input images that are later used. These methods were also implemented and the details of this implementation are to be found later in the text. In the last part the suggested methods of blood vessel detection are tested and the results evaluated.
Server Application and Mobile Application for iOS Devices for Capturing and Evaluating Image Quality for Wildlife Identification
Šesták, Šimon ; Šátek, Václav (referee) ; Rydlo, Štěpán (advisor)
The diploma thesis deals with the design of the application and REST API server and their implementations. The application is developed for the iOS operating system. Functionality of application should ensure simple communication with the server that will store and manage data. Since the server also stores photos of animals intended for identification, the application analyzes the image before uploading it to the server and sends it only if it contains the necessary information. A trained neural network model is used to analyze the image.
Meibograf Image Processing and Detection of Sebaceous/Meibomian Glands
Sandanus, Michal ; Malinka, Kamil (referee) ; Rydlo, Štěpán (advisor)
This thesis describes meibomian glands, methods of their observation and imaging. The current state-of-the-art in the field of segmentation of the eyelid and meibomian glands is presented, and based on it, a functional prototype of the application for solving the given problem using the methods of convolutional neural networks is created. Segmentation is performed by a 19-layer adaptation of the U-Net architecture and the MobileNetV3-large architecture with the LR-ASPP segmentation head. By combining these architectures, values that exceed the results of related works in certain metrics were achieved. The application is implemented as a desktop application in .NET 7.0 framework using OpenCV library.
Client Server Application for the Creation of File Pages for the PČR
Terbr, Filip ; Beran, Vítězslav (referee) ; Rydlo, Štěpán (advisor)
The diploma thesis is devoted to the design and implementation of a system for editing photographic documentation of the Police of the Czech Republic. The work includes an analysis of currently used technologies, a design of the client application, a design of the server part of the system and a description of the implementation of the client and server parts of the system. The resulting implementation of the server part is written in JavaScript with support for the Express.JS framework, the resulting client application is written in JavaScript using the Electron framework.
Smart Camera System for Museum Exhibitions
Homola, Radim ; Orság, Filip (referee) ; Rydlo, Štěpán (advisor)
This thesis focuses on development and construction of a camera system, that captures and post-processes video footage within museum environment in real time. Defined issue is resolved by separating the whole camera system into camera and computing unit, selecting correct components, mainly the camera module, and providing sufficient cooling for the computing unit. Following the solutions steps above enables to hide the powerful computing unit away from the visitor’s sight and also miniaturization of the visibly positioned camera unit. The developed camera system allows isolated operation of various algorithms, where the processed results can help exhibitors with better showpiece placements, recognition of unwanted behavior or protection from vandalism.
Presentation Attack Detection on Fingerprint Images Using Veins
Ondrušek, Tomáš ; Rydlo, Štěpán (referee) ; Kanich, Ondřej (advisor)
This work deals with the detection of a presentation attack on a biometric fingerprint sensor capturing an image of a vein. It contains a proposal for a method of detecting a presentation attack on a sensor that uses NIR illumination to highlight a vein in a finger. The method is tested and trained on a dataset containing 294 finger images extracted from 143 pictures of hands. The method is based on extracting textural information from the image and subsequent classification using the SVM classifier. The extraction uses parallel processing and processes features from 294 fingerprint images in 25 minutes on average. Subsequent classification achieves an average detection success rate of 97 %. The work also describes the steps necessary to carry out various types of attacks on a biometric system using the vein system.
Application for the Creation of File Pages for the PČR
Kleisl, Jan ; Sakin, Martin (referee) ; Rydlo, Štěpán (advisor)
The bachelor's thesis deals with the design and implementation of a file pages application for the Police of the Czech Republic. The goal of the thesis is to create a user-friendly and intuitive desktop application, that will allow more easily create and edit file pages due to software specifically made for that purpose. To achieve these goals, a survey was initially conducted to map mentioned area, requirements and needs of future users - police officers. The next step was to create a graphic user interface design and then implement it. The solution developed is a fully functional application that meets all of the assignment's requirements as well as the needs of the police. The main contribution of the thesis is to make activities of the Police of the Czech Republic more efficient.

National Repository of Grey Literature : 22 records found   1 - 10nextend  jump to record:
Interested in being notified about new results for this query?
Subscribe to the RSS feed.