National Repository of Grey Literature 2 records found  Search took 0.00 seconds. 
Using wavelet packet transformation for image compression
Kučera, Michal ; Rajmic, Pavel (referee) ; Kyselý, František (advisor)
The need for storing and transferring digital image data is still growing nowadays. The Compression is necessary to achieve effective usage of transfer and storage capacity. This Bachelor’s Thesis is concerned with compression method based on wavelet packet transform, which is derived from wavelet transform. It is focused especially on best basis search algorithms from whole wavelet packet tree. There is a comparison between 6 best basis and near-best basis search criterions by R. R. Coifman, M. V. Wickerhauser and C. Taswell in this thesis, which is achieved by using Matlab environment based program. The program was created for testing and demonstrative purpose only and that’s why it contains certain limitation such as only black and white image processing and its limited resolution. There is applied constant threshold on testing image after the several best basis criterions application. The Mean Squared Error is used for comparing outcome quality of compressed and original image. The function that acquires minimal Mean Squared Error is considered the best best-basis search criterion in term of visual quality. The results show that some of the Taswell’s functions significantly improve the visual quality of the image at the price of worse compression ratio against common Shannon entropy.
Using wavelet packet transformation for image compression
Kučera, Michal ; Rajmic, Pavel (referee) ; Kyselý, František (advisor)
The need for storing and transferring digital image data is still growing nowadays. The Compression is necessary to achieve effective usage of transfer and storage capacity. This Bachelor’s Thesis is concerned with compression method based on wavelet packet transform, which is derived from wavelet transform. It is focused especially on best basis search algorithms from whole wavelet packet tree. There is a comparison between 6 best basis and near-best basis search criterions by R. R. Coifman, M. V. Wickerhauser and C. Taswell in this thesis, which is achieved by using Matlab environment based program. The program was created for testing and demonstrative purpose only and that’s why it contains certain limitation such as only black and white image processing and its limited resolution. There is applied constant threshold on testing image after the several best basis criterions application. The Mean Squared Error is used for comparing outcome quality of compressed and original image. The function that acquires minimal Mean Squared Error is considered the best best-basis search criterion in term of visual quality. The results show that some of the Taswell’s functions significantly improve the visual quality of the image at the price of worse compression ratio against common Shannon entropy.

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