Image Compression using Wavelet Packet Tree based on Threshold Entropy

G.K. Kharate, A.A. Ghatol, and P.P. Rege (India)


Adaptive threshold, Compression, Orthogonal Wavelet, Wavelet Packet Tree, and Signal to Noise Ratio (SNR).


The paper describes a new technique for still colour image compression. The image compression includes transform of image, quantization and encoding. This paper mainly describes the transform of image and quantization technique. The algorithm is based on decomposition on orthogonal wavelet basis and it includes threshold entropy technique to construct the best tree. Quantization is adaptive, threshold is calculated on the basis of nature of image and type of wavelet is used. This paper also describes the comparison of wavelet transform and wavelet packet tree. The experimental aim is to determine the best method to improve the performance. The results are compared in terms of percentage of zeros; percentage of energy retained and signal to noise ratio for standard natural images as well as line-based images.

Important Links:

Go Back