|
|
Deep learning methods for image processing
Křenek, Jakub ; Chmelík, Jiří (referee) ; Kolář, Radim (advisor)
This master‘s thesis deals with the Deep Learning methods for image recognition tasks from the first methods to the modern ones. The main focus is on convolutional neural nets based models for classification, detection and image segmentation. These methods are used for practical implemetation – counting passing cars on video from traffic camera. After several test of available models, the YOLOv2 architecture was chosen and retrained on own dataset. The application also includes the addition of the SORT tracking algorithm.
|
|
|
Deep learning methods for image processing
Křenek, Jakub ; Chmelík, Jiří (referee) ; Kolář, Radim (advisor)
This master‘s thesis deals with the Deep Learning methods for image recognition tasks from the first methods to the modern ones. The main focus is on convolutional neural nets based models for classification, detection and image segmentation. These methods are used for practical implemetation – counting passing cars on video from traffic camera. After several test of available models, the YOLOv2 architecture was chosen and retrained on own dataset. The application also includes the addition of the SORT tracking algorithm.
|
|
|
Object detection and tracking
Fejfar, Petr ; Obdržálek, David (advisor) ; Bureš, Tomáš (referee)
This work focuses on object recognition and tracking by using laser range finder sensor. The topic examines various solutions proposed by other authors and analyzes the issue as a whole. The application of the task then takes place in a robot competition, which is used as a reference problem and solved subsequently. Port solutions are emphasized over PC platform solutions. The basis of the solution utilizes the principle of Kalman filters and the identification of objects by their position and speed. Powered by TCPDF (www.tcpdf.org)
|
|
|
Object detection and tracking
Fejfar, Petr ; Obdržálek, David (advisor) ; Plch, Tomáš (referee)
This paper focuses on object detection and tracking with the help of laser range finder sensor. The paper investigates other author's way of solving similar problems and analysis the problem as it is. The application of the task of robot competition is chosen as the reference problem which is solved as follows. The possible port solution for the non-PC platforms is emphasized. The basis of the solution is built on the idea of Kalman's filters and the identification of the objects acording to their position and speed. Powered by TCPDF (www.tcpdf.org)
|
guest ::
