Evaluation of factors affecting the selection of mother wavelet to improve image compression of Artificial & Natural Images
Nikkoo N. Khalsa, Gajanan Govind Sarate, D. T. Ingole · 2010
Image compression is essential for applications such as transmission and storage of databases. Artificial and Natural images are inherently voluminous. Therefore, efficient data compression techniques are essential for their archival and transmission of data. The standards use Discrete Cosine Transform (DCT) to reduce the spatial redundancy present in the images or video frames. The DCT has the main drawback of blocking artifacts. In recent time, wavelet transform (WT) has emerged as a popular technique for image compression applications. The wavelet transform has high de-correlation and energy compaction efficiency. The blocking artifacts is absent in a wavelet-based coder due to their overlapping basis functions. The wavelet transform can be composed of any function that satisfies requirements of multi-resolution analysis. In waveletbased image compression, the compression performance depends on the choice of wavelets. A review of the wavelet family used for compression is given in the literature. We have examined and compared various wavelet families such as Haar, Daubechies, Symlets, Coiflets, Biorthogonal and Reverse Biorthogonal using artificial and natural images. The results are measured in terms of PSNR considering as subjective quality measures.