The usage of perceptual measure Picture Quality Scale with neural network to evaluate local distortions in walvelet compressed images
Artykuł w czasopiśmie
| Status: | |
| Autorzy: | Skublewska-Paszkowska Maria |
| Rok wydania: | 2008 |
| Wersja dokumentu: | Drukowana | Elektroniczna |
| Język: | angielski |
| Numer czasopisma: | 3B |
| Wolumen/Tom: | 17 |
| Strony: | 408 - 412 |
| Efekt badań statutowych | NIE |
| Materiał konferencyjny: | NIE |
| Publikacja OA: | NIE |
| Abstrakty: | angielski |
| The perceptual measure Picture Quality Scale (PQS) together with the neural network were implemented to evaluate the local distortions in compressed images. The images were compressed with Daubechies wavelei transforms for degrees: 4.6.8.14, 20 and Embedded Zerotree Wavelet (EZW) coding. The neural network was learnt on the basis of visual evaluarion of a group of observers. For each pair of the original and the compressed images a sliding window of a fixed size was created. For this window the PQS yalue was computed and sayed as a new yalue to the image diat represented the local perceptual distorions. The PQS value was represented by the grey scale in the resultant image.rhe described usage of the PQS network enables the effective assessments based on Mean Opinion Score (MOS) measure so that the local distortions for compressed images can be created without observers. |
