|
|
Synthetic Dataset Generator for Traffic Analysis
Svoreň, Ondrej ; Sochor, Jakub (referee) ; Herout, Adam (advisor)
This bachelor thesis deals with the creation and customization of synthetic dataset genera tor for traffic analysis. It focuses on traffic analysis by means of computer vision, methods and conditions of creating the generator of synthetic dataset, possible application of achie ved results in machine learning and additional development opportunities. Using available automobile photographs from the Czech Republic, Slovakia, Poland and Hungary, a synthe tic license plate number generator was created, which, after graphical adjustment and after joining with the vehicle photographs creates the resulting dataset for machine learning. The solution itself is divided into the three scripts written in Python using the OpenCV library. The resulting dataset serves as an input for the machine learning system to re-identify the license plate numbers from photographs captured in the flow of traffic.
|
|
|
Detection of Weapons in 2D Image
Demčák, Ján ; Spurný, Martin (referee) ; Drahanský, Martin (advisor)
This bachelor thesis deals with detection of weapons in 2D image. In the theoretical part of the thesis the term weapon was defined and the possibilities of detection of weapons in image with using classic methods and deep neural networks were mentioned there. The key steps of image processing, objects classification and detection were described. The overview of frameworks, libraries was presented. To implement the pratical part of the thesis, 3 models were chosen. The first classic model with using HOG transformation. The second CNN model with priority target detection accuracy and with two different neural network architectures as classifiers. The third model with YOLO network architecture had as priority target real-time detection. The essential part of each model was choice, or more precisely creating suitable dataset. What followed was the construction and implementation of models and the evaluation of obtained data.
|
|
|
Widget Recognition for Visual GUI Testing
Kisela, Sebastián ; Peringer, Petr (referee) ; Smrčka, Aleš (advisor)
The bachelor thesis analyzes current methods and tools used for user interfaces testing. The thesis sets specification requirements and design of the implemented tool. Implementation and limitations of the implemented tool, which is based on visual object recognition, are explained in a detailed way. Verification process of the tool is described at the end.
|
|
|
Applications of Ant Algorithms
Kaščák, Imrich ; Vašíček, Zdeněk (referee) ; Bidlo, Michal (advisor)
The presented thesis focuses on the basic optimization algorithm of Ant Colony Optimization (ACO) - Ant System (AS) and its extension, Ant Colony System (ACS) on Traveling Salesman Problem (TSP). The essence of these algorithms is to find the optimal solution (the shortest path) in a specified instance containing several locations. The thesis demonstrates verification of behaviour of both algorithms, experimental study of impact of adjustable parameters of ant algorithms on result. Furthermore, the thesis is focused on examining the idea of optimizing detection of image edges by introducing a modification into an existing solution. Modified solution experiments are performed and compared to the original.
|
|
|
Convolutional Neural Networks
Lietavcová, Zuzana ; Zbořil, František (referee) ; Zbořil, František (advisor)
This thesis deals with convolutional neural networks. It is a kind of deep neural networks that are presently widely used mainly for image recognition and natural language processing. The thesis describes specifics of convolutional neural networks in comparison with traditional neural networks and is focused on inner computations in the process of learning. Convolutional neural networks typically consist of a different types of layers of neurons and the core part of this thesis is to demonstrate computations of individual types of layers. Learning demonstrating program of a simple convolutional network was designed and implemented using own implementation of neural network. Validity of the implementation was tested by training models for solving a classification task. Experiments with different types of architectures were conducted and their performance was compared.
|
|
|
Machine vision for locating objects
Hasík, Dávid ; Huzlík, Rostislav (referee) ; Vetiška, Jan (advisor)
The bachelor’s thesis deals with design of machine vision system for object localization and its application possibilities. Design approach, its components and their individual advantages and disadvantages for this type of machine vision system are described. This thesis also includes simple model of machine vision system for object localization using regular PC. For image processing is used Vision Assistant and its tools. Two scripts are made in this software and their performance is compared.
|
|
|
Gamification of line-follower robot videorecordings
Soboňa, Tomáš ; Honec, Peter (referee) ; Janáková, Ilona (advisor)
This bacheor’s thesis is concerned with the processing of a vehicle’s trajectory following a black line located on an unchanging white background, and subsequently comparing of the quality of each vehicle’s covered distance on this trajectory. The first part of the thesis describes the theory for recognizing the track, individual graphic tags, as well as detection of important points in the picture and their correlation with points in another picture, for which an algorithm SIFT is used. The second part deals with the description of the program, which was created as part of this work, its classes and methods. The program is written in Python, for image processing is used mainly an open source library OpenCV and, to a lesser extent, the NumPy library. Finally, the results are concluded at the end of the thesis.
|
|
|
Disease Detection in Eye Retina Image
Koštialik, Daniel ; Semerád, Lukáš (referee) ; Maruniak, Lukáš (advisor)
Diabetic retinopathy and age related macular degeneration are among the most common eye retina diseases, which cause partial or complete blindness. The main goal of this thesis is to design and implement software for automatic detection of symptoms from eye fundus images. The detection algorithm is based on the image segmentation by region growing method and afterwards analysis. Determination of retina objects such as optic disc, macula and blood vessels is important prior symptoms detection as they can adversely affect the results of the analysis. Total 259 images were analysed and algorithm reaches more than 90 % average success rate. The algorithm, in combination with appropriate hardware and optic mechanism, forms one of practical application in global population screening. Thanks the automatic detection it is possible to determine the presence of symptoms and start an early treatment.
|
|
|
Applications for image processing for Android OS
Kiac, Martin ; Říha, Kamil (referee) ; Krajsa, Ondřej (advisor)
This work deals with an issue of image processing and detection of predefined objects from the camera on Android platform. This work uses image processing tools and methods from OpenCV library. The work consists of a theoretical part, in which the Android platform itself is described, the structure of the operating system and Android Studio development environment. The theoretical part also contains a description of OpenCV library and describes the theory of image processing using this library. The practical part of this work describes implementation of OpenCV library to development environment Android Studio. The practical part also describes techniques for image processing by OpenCV library. Subsequently this work contains analysis and searching for contours, the use of Hough transformation and segmentation of the processed image. This work also describes realization of graphical user interface and subsequently work with the application's database. The conclusion is about final evaluation of the work and achieved results.
|
|
|
Object detection
Baáš, Filip ; Petyovský, Petr (referee) ; Richter, Miloslav (advisor)
This bachelor thesis deals with detection of rigid objects in images. Chamfer matching algorithm, which is built for this kind of tasks is used as detection algorithm. First part of this work is dedicated to theoretical explanation of the algorithm. Most commonly used metrics of distance transform are explained, which is needed for the algorithm. Also explanation of chamfer distance calculation and pyramid representation of information is here. Next part is dedicated to development tools used in this work, which is integrated development environment Visual Studio and libraries OpenCV for image processing and Qt for graphical user interface creation. In last part of this work, practical implementation of object detection is described. This part explains the way objects are rendered, steps for creating a template from rendered image, method to create set of templates, comparison of speed of distance transformation calculation in different metrics, comparison of speed of common and pyramid detection and method of score calculation. The conclusion summarizes reached goals of this work.
|
guest ::
