|
|
Automatic Construction of a Terrain Map by a Drone
Kapsa, Jan ; Bambušek, Daniel (referee) ; Beran, Vítězslav (advisor)
This thesis focuses on how UAV mapping functions with the focus on methods working in real-time. The process of image stitching is thoroughly explained and 2 methods based on it are designed. Together with these methods dataset is built containing different situations. Maps of the terrain are created by these methods, which are then compared and scored.
|
|
|
Složení snímku prstu s viditelným krevním řečištěm.
Pekárek, Jakub ; Semerád, Lukáš (referee) ; Rydlo, Štěpán (advisor)
This thesis describes a method for creating a composite image with a visible bloodstream. The bloodstream is detected by the Maximum Curvature algorithm. After that it was experimented with detecting and matching keypoints via algorithms SIFT and ORB. However, the resulting matches from these algoritms were not good enough, therefore the keypoints have to be manually assigned. From these keypoints, an estimated tranformation matrix was created. This matrix was then used for connecting two images. A cycle of assigning keypoints and connecting images is repeated even for newly created images. The final composite image with a visible bloodstream is created after connecting last two images. This composite image can be further used for verification or identification people.
|
|
|
Deep Learning for Image Stitching
Šilling, Petr ; Beran, Vítězslav (referee) ; Španěl, Michal (advisor)
Sešívání obrázků je klíčovou technikou pro rekonstrukci objemů biologických vzorků z překrývajících se snímků z elektronové mikroskopie (EM). Současné metody zpracování snímků z EM k sešívání zpravidla využívají ručně definované příznaky, produkované například technikou SIFT. Nedávný vývoj však ukazuje, že konvoluční neuronové sítě dokáží zlepšit přesnost sešívání tím, že se naučí diskriminativní příznaky přímo z trénovacích obrázků. S ohledem na potenciál konvolučních neuronových sítí tato práce navrhuje sešívací nástroj DEMIS, který staví na pozornostní síti LoFTR pro hledání shodných příznaků mezi páry obrázků. Dále práce navrhuje novou datovou sadu generovanou dělením obrázků z EM s vysokým rozlišením na pole překrývajících se dlaždic. Výsledná datová sada je použita pro dotrénování sítě LoFTR a k vyhodnocení nástroje DEMIS. Experimenty na dané datové sadě ukazují, že nástroj je schopen nalézt přesnější shody mezi příznaky než SIFT. Navazující experimenty na obrázcích s vysokým rozlišením a malými překryvy mezi dlaždicemi dále poukazují na výrazně vyšší robustnost oproti metodě SIFT. Dosažené výsledky celkově naznačují, že hluboké učení může vést k prospěšným změnám v oblasti EM, například k umožnění menších překryvů mezi snímanými obrázky.
|
|
|
Object tracking in video
Boszorád, Matej ; Přinosil, Jiří (referee) ; Rajnoha, Martin (advisor)
This bachelor thesis deals with the issue of tracking multiple objects in a video, specifically focusing on non-learning algorithms. The first chapter represents the theoretical part of the thesis, in which some of the often used tracking methods are described, such as mean-shift, scale-invariant object transformation, Kalman filter, particle filter and Gabor wavelet transformation. These algorithms are broken down by properties they use for proper tracking. The chapter also contains section assignment problem, which is mainly concerned with Hungarian algorithm. The next part describes options of merging multiple tracking methods that are broken down by construction type into parallel, cascade, weighted and discriminatory with example for each one. Moreover there is described adaptability of the tracking system. Bellow are described problems which may occur during tracking and possible solutions to them. This section consists of a solution of image noise, changes in illumination, appearance and extinction of an object, focusing mainly on solving the problem of object occlusion. Within the practical part is created algorithm composed of different types of tracking, the results of which are then compared with selected tracking systems from the multiple object tracking benchmark. The practical part includes the tools used and the explanation of the design, in which the main classes and methods used for the tracking are explained. Besides that, this section describes parallel merging and tracking adaptability . The results of the thesis contain a comparison of the use of tracking techniques separately and together. To compare the results, videos for pedestrian tracking and face tracking were used. This thesis was based on the assumption that merging multiple monitoring systems will help with the improvement of the tracking, which was confirmed by the results.
|
|
| |
|
|
Robot Localization Using Camera
Heřman, Petr ; Španěl, Michal (referee) ; Beran, Vítězslav (advisor)
The objective of this work is to design a simple localization method and its implementation in robot operating system ROS. This method uses a monocular camera as the only sensor and estimates the position in a known map. In experiments with prototypes are tested key points of type SURF, SIFT and ORB.
|
|
|
Object Detection and Tracking Using Interest Points
Bílý, Vojtěch ; Hradiš, Michal (referee) ; Juránek, Roman (advisor)
This paper deals with object detection and tracking using iterest points. Existing approaches are described here. Inovated method based on Generalized Hough transform and iterative Hough-space searching is proposed in this paper. Generality of proposed detector is shown in various types of objects. Object tracking is designed as frame by frame detection.
|
|
|
Panoramatic View Reconstruction
Kuzdas, Oldřich ; Kohoutek, Michal (referee) ; Říha, Kamil (advisor)
This paper deals step by step with process of stitching images taken by perspective camera rotated by its optical center into the panoramic image. There are described keypoint searching algorhytms, possibilities of calculating homography matrix and methods of eliminating unwanted seams between source images in final panoramic image. A part of this paper is also standalone application in which are implemented some algorhytms described in the work.
|
|
|
Perimeter Monitoring and Intrusion Detection Based on Camera Surveillance
Goldmann, Tomáš ; Drahanský, Martin (referee) ; Orság, Filip (advisor)
This bachelor thesis contains a description of the basic system for perimeter monitoring. The main part of the thesis introduces the methods of computer vision suitable for detection and classification of objects. Furthermore, I devised an algorithm based on background subtraction which uses a Histogram of Oriented Gradients for description of objects and an SVM classifier for their classification. The final part of the thesis consists of a comparison of the descriptor based on the Histogram of Oriented Gradients and the SIFT descriptor and an evaluation of precision of the detection algorithm.
|
|
|
3D model
Sládeček, Martin ; Petyovský, Petr (referee) ; Richter, Miloslav (advisor)
This paper concerns the task three-dimensional scene using image sequence obtained with an ordinry camera. First portion of this thesis outlines the principles used in solving of the task, the second chapter describes a reconstucion algorithm and it's implementation in the Python programming language. The output of this program is demonstrated on several selected scenes. Final remarks discuss the quality of resulting models, shortcomings of the program and possible improvements.
|
guest ::
