|
|
Suppresion of distortion in iris images
Jalůvková, Lenka ; Štohanzlová, Petra (referee) ; Kolář, Radim (advisor)
This master`s thesis is focused on a suppression of a distorsion in iris images. The aim of this work is to study and describe existing degradation methods (1D motion blur, uniform 2D motion blur, Gaussian blur, atmospheric turbulence blur, and out of focus blur). Furthermore, these methods are implemented and tested on a set of images. Then, we designed methods for suppression of these distorsions - inverse filtration, Wiener filtration and iterative deconvolution. All of these methods were tested and evaluated. Based on the experimental results, we can conclude that the Wiener-filter restoration is the most accurate approach from our test set. It achieves the best results in both normal and iterative mode.
|
|
|
Optic disc detection in retinal images
Jalůvková, Lenka ; Odstrčilík, Jan (referee) ; Kolář, Radim (advisor)
The bachelor thesis is focused on a detection of optic disc in the retinal images in order to propose and compare several existing methods. The detection is implemented as the Gaussian filter, matched filter and is done by vascular structure information. The DIARETDB1 database is used for testing. The best results have been achieved using Gaussian filter and detection by vascular structure information with success rate 81%. The description and comparison of all the algorithms can be found in this thesis.
|
|
|
Live Panorama
Jalůvková, Lenka ; Juránková, Markéta (referee) ; Herout, Adam (advisor)
This batchelor`s thesis is focused on creation of live panorama. This is an innovative representation of data from streaming cameras. The input is a video from the streaming camera and the output is also stream - constantly re-drawing panorama according to the current camera position. The resulting solution is based on the SIFT algorithm for finding keypoints in the frames. Furthermore, the RANSAC algorithm is used for calculating homography for stitching images together. Re-drawing the panorama requires weighting of edges. The functionality of the proposed solution was tested on a series of testing video recordings.
|
|
|
Optic disc detection in retinal images
Jalůvková, Lenka ; Odstrčilík, Jan (referee) ; Kolář, Radim (advisor)
The bachelor thesis is focused on a detection of optic disc in the retinal images in order to propose and compare several existing methods. The detection is implemented as the Gaussian filter, matched filter and is done by vascular structure information. The DIARETDB1 database is used for testing. The best results have been achieved using Gaussian filter and detection by vascular structure information with success rate 81%. The description and comparison of all the algorithms can be found in this thesis.
|
|
|
Live Panorama
Jalůvková, Lenka ; Juránková, Markéta (referee) ; Herout, Adam (advisor)
This batchelor`s thesis is focused on creation of live panorama. This is an innovative representation of data from streaming cameras. The input is a video from the streaming camera and the output is also stream - constantly re-drawing panorama according to the current camera position. The resulting solution is based on the SIFT algorithm for finding keypoints in the frames. Furthermore, the RANSAC algorithm is used for calculating homography for stitching images together. Re-drawing the panorama requires weighting of edges. The functionality of the proposed solution was tested on a series of testing video recordings.
|
|
|
Suppresion of distortion in iris images
Jalůvková, Lenka ; Štohanzlová, Petra (referee) ; Kolář, Radim (advisor)
This master`s thesis is focused on a suppression of a distorsion in iris images. The aim of this work is to study and describe existing degradation methods (1D motion blur, uniform 2D motion blur, Gaussian blur, atmospheric turbulence blur, and out of focus blur). Furthermore, these methods are implemented and tested on a set of images. Then, we designed methods for suppression of these distorsions - inverse filtration, Wiener filtration and iterative deconvolution. All of these methods were tested and evaluated. Based on the experimental results, we can conclude that the Wiener-filter restoration is the most accurate approach from our test set. It achieves the best results in both normal and iterative mode.
|
guest ::
