|
|
object matching
Mišta, Petr ; Petyovský, Petr (referee) ; Richter, Miloslav (advisor)
Detection of objects based on color is not commonly used method of computer vision. There are many methods thats deals with the detection of significant points, but the color information has been omitted. The goal of this thesis is to design method of the detection significant color image areas and these areas match up with areas detected in another image. I analyzed features of detectors required to identify the reciprocal correspondence of images, defined the color significance concept, described basic color models and their properties, and a design of statistically compiled data - based method was described. Algorithms for the detection of color use color models RGB and HSV. Correspondence of areas detected in different images is performedy Kohonen neural network. The first input vector can teach Kohonen neural network and the second vector is classified by this network. To remove erroneous classifications RANSAC method is used. As a result, the method can be used for basic and rapid determination of correspondence between images, or to speed up commonly used methods for detection of significant points. At the end of the thesis are presented programs, showing functionality and options of design methods. The designed algorithms have been developed in MATLAB.
|
|
|
Lossless Image Compression
Komjáthy, Gergely ; Polok, Lukáš (referee) ; Bařina, David (advisor)
This thesis deals with lossless image compression. In this paper are shown some colour models, which can be used for lossless image compression and formulas how to convert them to RGB and vica versa. You can learn predictors, how they work and discription of some of them. There is described the function of arithmetic coder, PPM coder and a brief description of Huffman coding.
|
|
|
Lossless Image Compression
Vondrášek, Petr ; Polok, Lukáš (referee) ; Bařina, David (advisor)
The aim of this master's thesis was to design, develop and test a method for lossless image compression. The theoretical part includes a description of selected exiting methods such as RLE, MTF, adaptive arithmetic coding, color models used in LOCO-I and JPEG 2000, predictors MED, GAP and laplacian pyramid. The conclusion includes a comparison of various combinations of chosen approaches and overall efficiency compared with PNG and JPEG-LS.
|
|
|
Impact of color models on performance of convolutional neural networks
Šimunský, Martin ; Doležel, Petr (referee) ; Škrabánek, Pavel (advisor)
Current knowledge about impact of colour models on performance of convolutional neural network is investigated in the first part of this thesis. The experiment based on obtained knowledge is conducted in the second part. Six colour models HSV, CIE 1931 XYZ, CIE 1976 L*a*b*, YIQ a YCbCr and deep convolutional neural network ResNet-101 are used. RGB colour model achieved the highest classification accuracy, whereas HSV color model has the lowest accuracy in this experiment.
|
|
|
Color correction of 3D model based on scene knowledge and light illumination
Dočkal, Michal ; Kopečný, Lukáš (referee) ; Chromý, Adam (advisor)
This thesis focus is on the design and verification of improvements for the RoScan robotic system texture module. RoScan is a robotic scanner for scanning and diagnosing the human body. In this thesis current texturing module of this system is described. Furthermore this thesis describes the theory of light, color and shading in computer graphics. Subsequently, improvments for RoScan texture module are proposed based on the said principles. The last part deals with the implementation of the test script in Matlab and the verification of the functionality of the proposed solution.
|
|
|
Colorimeter
Škoda, Pavel ; Janáková, Ilona (referee) ; Honec, Peter (advisor)
This thesis deals with the analysis of the color parameters for their subsequent use in the design of the device, which will be used to measure their stability in continuous production. The theoretical part defines the basic concepts and knowledge needed to understand the problems. In addition, there are described individual components and sensors suitable for color measurement. The practical part deals mainly with the descrip- tion of the construction of the colorimeter, its individual parts and the creation of the control program. At the end, there are results of the testing of the product and their overall assessment
|
|
|
Lossless Video Compression
Polák, Michael Adam ; Polok, Lukáš (referee) ; Bařina, David (advisor)
This thesis is describing methods of lossless video compression. It also contains the description and formulas for conversions of color models suitable for lossless compression. The last theoretical part describes context compression and entropy coding with emphasis on arithmetic coding. Following the arithmetic coder is the description and details about the implemented video codec. This codec is then tested and compared with the currently used FFV1 and Huffyuv codecs. The result of this thesis is that adding the number of contexts improves the compression ratio although it does not outperform the compression ratio of FFV1, the compression ratio of Huffuyv was always outperformed.
|
|
|
Image based road surface segmentation with lane detection
Turoň, Rudolf ; Honec, Peter (referee) ; Zemčík, Tomáš (advisor)
This bachelor thesis deals with road semantic segmentation with lane lines detection. The aim of the thesis is to conduct a survey of methods used for general segmentation and lane lines detection, collect a representative dataset for testing, design a data processing system including implementation for semantic segmentation and detection of lanes in image, video and live camera input. In preprocessing phase images are filtered using morphological transformations. The segmentation is performed using the watershed method with labels, and an adaptive algorithm is designed to find the labels. The lanes are searched for in the thresholded projectively transformed image using the sliding window technique. As a result, a segmentation accuracy of 88.3% is achieved based on the IoU metric. The thesis concludes with a discussion of the obtained results and summarizes the possibilities for further improvements of the system.
|
|
|
Interactive web applications for image processing education
Krolop, Filip ; Záviška, Pavel (referee) ; Rajmic, Pavel (advisor)
Introduction to theory related to applets that will be used for educational purposes, creation of custom design, draft of their basic functionality and control elements, followed by own implementation and optimization. This is an interactive submission of image processing algorithms specifically, applets are focused on color models, their relationships and calculations between each other: 1 / Pallet (index) image representation, 2 / Colormixing in different models, 3 / Color image conversion to grayscale image, 4 / Conversion of a grayscale image to color image (false colors). Applets implemented using HTMLand JavaScript to demonstrate the theory and encourage students to interact.
|
|
|
Education Computer Program for Demonstration of Colors and Colors Models Principles
Vojíř, Jaromír ; Jančík, Zdeněk (referee) ; Kršek, Přemysl (advisor)
The thesis deals with development of an educational program for demonstration of principles of colors and of color models. The thesis contains theoretical study of the given question from basic terms up to use of individual color models. This is followed by a detailed analysis of the requirements necessary for development of the given educational program. In the implementation part, the process of development of the application itself is described. In the end of the thesis there is a discussion of contribution of the thesis and of possibility of future adaptations. The program can serve as a complement of study materials for students.
|
guest ::
