Dynamic Selection of Suitable Wavelet for Effective Color Image Compression using Neural Networks and Modified RLC
Keywords:
image compression, dynamic selection, wavelet, counter propagation neural network, MLFFNN, EBP
Abstract
Image Compression has become extremely important today with the continuous development of internet, remote sensing and satellite communication techniques. In general, single Wavelet is not suitable for all types of images. This paper proposes a novel approach for dynamic selection of suitable wavelet and effective Image Compression. Dynamic selection of suitable wavelet for different types of images, like natural images, synthetic images, medical images and etc, is done using Counter Propagation Neural Network which consists of two layers: Unsupervised Kohonen (SOFM) and Supervised Gross berg layers. Selection of suitable wavelet is done by measuring some of the statistical parameters of image, like Image Activity Measure (IAM) and Spatial Frequency (SF), as they are strongly correlated with each other. After selecting suitable wavelet, effective image compression is done with MLFFNN with EBP training algorithm for LL2 component. Modified run length coding is applied on LH2 and HL2components with hard threshold and discarding all other sub-bands which do not effect much the quality (both subjective and objective) (HH2, LH1, HL1 and HH1). Highest CR (191.53), PSNR (78.38 dB), and minimum MSE (0.00094) of still color images are obtained compared to SOFM, EZW and SPIHT.
Downloads
- Article PDF
- TEI XML Kaleidoscope (download in zip)* (Beta by AI)
- Lens* NISO JATS XML (Beta by AI)
- HTML Kaleidoscope* (Beta by AI)
- DBK XML Kaleidoscope (download in zip)* (Beta by AI)
- LaTeX pdf Kaleidoscope* (Beta by AI)
- EPUB Kaleidoscope* (Beta by AI)
- MD Kaleidoscope* (Beta by AI)
- FO Kaleidoscope* (Beta by AI)
- BIB Kaleidoscope* (Beta by AI)
- LaTeX Kaleidoscope* (Beta by AI)
How to Cite
Published
2014-03-15
Issue
Section
License
Copyright (c) 2014 Authors and Global Journals Private Limited
This work is licensed under a Creative Commons Attribution 4.0 International License.