|
|
Detection, Localization, and Decoding QR Code in Image
Kałuża, Marian ; Juránková, Markéta (referee) ; Herout, Adam (advisor)
This thesis describes the QR Code detector based on detecting edge lines using Hough Transform and point-to-line parametrization. Generally, the presented method can be also used for detecting different types of matrix codes. This approach searches in the accumulated Hough space for evidence of the matrix code as one compound object. It detects parallel lines by finding a concurrent lines' image groups in the transformed space. It contrasts with the common approach which is first detect the lines and calculate the vanishing point afterwards, which omit much of the useful information stored in the Hough space which can be useful for the detection of vanishing points. The purpose of this thesis is a research of methods which can detect the QR Code in spite of rotation, perspective distortion or uneven illumination. The experimental evaluation shows that the presented algorithm is stable and accurate and is tolerant to high degrees of perspective deformation.
|
|
|
Camera calibration
Opravil, Jan ; Babinec, Tomáš (referee) ; Richter, Miloslav (advisor)
Imperfection of some lenses used in industrial cameras is distracting. This thesis is concerned with the calibration of cameras with geometric barrel or cushiontype distortion. In the calibrating process is used a function which approximates real lens distortion. To calculate the coefficients of this function the straight line sections in the distorted image are used. Those detected lines play the key role and they ought to have the greatest possible information bias. To enhance the quality of detection are chosen some methods of preprocessing. The algorithms thus created are tested first on an artificially generated calibers and later on real ones. The program is written in C and is using some of the functions of the OpenCV library.
|
|
|
Hough Transform and its Variants
Stejskal, Jan ; Přinosil, Jiří (referee) ; Říha, Kamil (advisor)
This thesis focuses on basic principles of Hough transform for the detection of lines, circles and arbitrary shapes, their algorithms, computational complexity and preparation of a picture for detection. The description of implementation of individual algorithms and creation of an aplication that demonstrates their properties. Implementation has been done in programming language MS Visual C++ with the help of open source library OpenCV and user interface has been created using IDE Qt Creator 4.13.2 (Community).
|
|
|
Uniform Marker Field on a Cylinder
Kříž, Radim ; Havel, Jiří (referee) ; Herout, Adam (advisor)
This work presents a new extension for Uniform Marker Field, which is able to detect UMF on the cylinder. First part of the text deals with Augmented reality and focuses on systems using markers. It discusses the actual state-of-the-art systems and its possibilities. After that it focuses more deeply on the marker system Uniform marker field and its grayscale variants. Next part of the work describes properties of the cylinder projected in real space. Important properties for detecting are discussed in detail. Then the proposal and description of detection algorithm is presented. Implementation of algorithm is tested and evaluated on the very end of this thesis.
|
|
|
Hough Transform and its Variants
Stejskal, Jan ; Přinosil, Jiří (referee) ; Říha, Kamil (advisor)
This thesis focuses on basic principles of Hough transform for the detection of lines, circles and arbitrary shapes, their algorithms, computational complexity and preparation of a picture for detection. The description of implementation of individual algorithms and creation of an aplication that demonstrates their properties. Implementation has been done in programming language MS Visual C++ with the help of open source library OpenCV and user interface has been created using IDE Qt Creator 4.13.2 (Community).
|
|
|
Detection, Localization, and Decoding QR Code in Image
Kałuża, Marian ; Juránková, Markéta (referee) ; Herout, Adam (advisor)
This thesis describes the QR Code detector based on detecting edge lines using Hough Transform and point-to-line parametrization. Generally, the presented method can be also used for detecting different types of matrix codes. This approach searches in the accumulated Hough space for evidence of the matrix code as one compound object. It detects parallel lines by finding a concurrent lines' image groups in the transformed space. It contrasts with the common approach which is first detect the lines and calculate the vanishing point afterwards, which omit much of the useful information stored in the Hough space which can be useful for the detection of vanishing points. The purpose of this thesis is a research of methods which can detect the QR Code in spite of rotation, perspective distortion or uneven illumination. The experimental evaluation shows that the presented algorithm is stable and accurate and is tolerant to high degrees of perspective deformation.
|
|
|
Uniform Marker Field on a Cylinder
Kříž, Radim ; Havel, Jiří (referee) ; Herout, Adam (advisor)
This work presents a new extension for Uniform Marker Field, which is able to detect UMF on the cylinder. First part of the text deals with Augmented reality and focuses on systems using markers. It discusses the actual state-of-the-art systems and its possibilities. After that it focuses more deeply on the marker system Uniform marker field and its grayscale variants. Next part of the work describes properties of the cylinder projected in real space. Important properties for detecting are discussed in detail. Then the proposal and description of detection algorithm is presented. Implementation of algorithm is tested and evaluated on the very end of this thesis.
|
|
|
Camera calibration
Opravil, Jan ; Babinec, Tomáš (referee) ; Richter, Miloslav (advisor)
Imperfection of some lenses used in industrial cameras is distracting. This thesis is concerned with the calibration of cameras with geometric barrel or cushiontype distortion. In the calibrating process is used a function which approximates real lens distortion. To calculate the coefficients of this function the straight line sections in the distorted image are used. Those detected lines play the key role and they ought to have the greatest possible information bias. To enhance the quality of detection are chosen some methods of preprocessing. The algorithms thus created are tested first on an artificially generated calibers and later on real ones. The program is written in C and is using some of the functions of the OpenCV library.
|
guest ::
