Embedded Zero-Tree Wavelet Coding with Selective Decomposition Bands

Wavelets and its applications became widespread with the inception of embedded zerotree wavelet coding since past two decades. The wavelet transform decomposes any signal into different decomposition bands. As the level of decomposition progresses, the bands are further decomposed. But, the low freq...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2019 International Conference on Communication and Signal Processing (ICCSP) S. 0445 - 0449
Hauptverfasser: Narasihimhaprasad, K. L., Nagabhushanam, M. V., Satyanarayana Tallapragada, V. V., Krishna Sunkara, Jaya
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.04.2019
Schlagworte:
Compression Ratio
Encoding
EZW
HH band
Image coding
Image reconstruction
Significant Map
Simulation
Wavelet Based Image Coding
Wavelet transforms
Zirconium
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Wavelets and its applications became widespread with the inception of embedded zerotree wavelet coding since past two decades. The wavelet transform decomposes any signal into different decomposition bands. As the level of decomposition progresses, the bands are further decomposed. But, the low frequency bands carry significant role in reconstruction. When a wavelet based coding scheme is to be applied, all the bands are being considered, which raises memory requirement on the encoded or compressed data. In this paper, EZW coding is exploited in depth to identify the ways of reducing number of symbols generated by which the Compression Ratio(CR) will be improved without reducing much of the quality. Now the proposal of the paper is that, as HH band constitutes a minor role in reconstruction, the band is supposed to be ignored, making all the coefficients in that band insignificant expecting a significant improvement in compression ratio and PSNR. Results show better PSNR and CR.
DOI:10.1109/ICCSP.2019.8698062