National Repository of Grey Literature 3 records found  Search took 0.00 seconds. 
Image Processing on GPUs
Horák, Lukáš ; Španěl, Michal (referee) ; Smrž, Pavel (advisor)
Acceleration of image processing can be done by more than one way. This work is oriented to new technology using graphics processing unit with architecture which allows universal usage. The architecture has called as NVIDIA CUDATM (Cumpute unified device architecture). This work instructing about this technology and show basic terminology needed to understand other chapters. Convolution image processing is selected to testing this technology. It describes some first mistakes in algorithm design and show other steps of acceleration.
Image Processing on GPUs
Horák, Lukáš ; Španěl, Michal (referee) ; Smrž, Pavel (advisor)
Acceleration of image processing can be done by more than one way. This work is oriented to new technology using graphics processing unit with architecture which allows universal usage. The architecture has called as NVIDIA CUDATM (Cumpute unified device architecture). This work instructing about this technology and show basic terminology needed to understand other chapters. Convolution image processing is selected to testing this technology. It describes some first mistakes in algorithm design and show other steps of acceleration.
Detection and Tracking of Small Moving Objects
Filip, Jan ; Zuzaňák, Jiří (referee) ; Hradiš, Michal (advisor)
Thesis deals with the detection and tracking of small moving objects from static images. This work shows a general overview of methods and approaches to detection and tracking of objects. There are also described some other approaches to the whole solution. It also included basic definitions, such a noise, convolution and mathematical morphology. The work described Bayesian filtering and Kalman filter. It described the theory of Wavelets, wavelets filters and transformations. The work deals with different ways of the blob`s detection. It is here the design and implementation of applications, which is based on the wavelets filters and Kalman filter. It`s implemented several methods of background subtraction, which are compared by testing. Testing and application are designed to detect vehicles, which are moving faraway (at least 200 m away). 

Interested in being notified about new results for this query?
Subscribe to the RSS feed.