Hybrid differential pulse code modulation in wavelet domain and lossless methods for image compression

Abstract EN

Digital images are widely used in computer applications.

Uncompressed digital images require considerable storage capacity and transmission bandwidth.

Efficient image compression solutions are becoming more critical with the recent growth of data intensive and multimedia-based web application.

In this paper, Hybrid Differential Pulse Code Modulation (DPCM) in wavelet domain and lossless methods (Huffman and Run length code) for image compression is investigated.

The image signal is composed into four scales (approximate, horizontal, vertical and diagonal coefficients).

The energy level of approximate is different from that of detail information ; therefore, the approximate is quantized alone and horizontal and vertical are quantized alone by using the differential pulse code modulation (DPCM).

In all simulations the diagonal coefficients are discarded.

Two types of coding are used, Huffman code and run length coding (RLC) techniques.

The results show that truncate horizontal and vertical coefficients using hard threshold gives best performance than soft threshold with gain about 3 dB to get the same compression ratio (CR).

The wavelet based image compression using Huffman code gives CR about 77 % compared with RLC that gives CR about 71 % with suitable peak signal to noise ratio (PSNR).