On the application of wavelet transform and Huffman algorithm to Yorùbá language syntax text files compression

Abstract

Most algorithms of data compression were developed with English language as target text syntax. However, this paper approaches the problem of Yorùbá text files compression via the use of Discrete Wavelet Transform (DWT) and Huffman algorithm. Text files in Yorùbá language syntax are first converted into signal format that are then decomposed using DWT. The decomposed ASCII code representation of the text files are subsequently encoded using Huffman algorithm. Twenty different variants of DWTs taken from four families of wavelet filters (Haar, Daubechies, Symlets and bi-orthogonal) are considered to select the optimal DWT for Yorùbá text files compression. Furthermore, experiments are carried out in the proposed compression scheme with six different Yorùbá text files extracted from the open sources as input data sets. It is found that out of the twenty variants of DWT investigated, sym6 gives the best output for effective Yorùbá text files compression, due to its relatively high compression ratio, high compression factor and lowest compression error. Thus, sym6 as a wavelet transform is suitable for lossy text compression algorithm meant for Yorùbá language syntax text files.

Keywords

• text file
• compression
• wavelet transform
• huffman coding
• yorùbá language syntax