Sensitivity analysis of daubechies 4 wavelet coefficients for reduction of reconstructed image error

  • Authors:
  • Devinder Kaur;Pat Marshall

  • Affiliations:
  • EECS, University of Toledo, OH;WPAFB, OH

  • Venue:
  • SIP'07 Proceedings of the 6th Conference on 6th WSEAS International Conference on Signal Processing - Volume 6
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Quantization of compressed image file reduces the requirements for memory storage and the transmission bandwidth. However, the cost of quantization is loss of information. On the receiver end when image is reconstructed from lossy de-quantized file using the wavelet reconstruction coefficients, it is impossible to create a perfect image exactly like the original image. This paper records the sensitivity analysis of wavelet coefficients to determine which coefficients contribute to the image reconstruction and in what way and how they should be changed to reconstruct a perfect image. Some interesting findings have been recorded in the conclusion section.