WAVELET PACKETS
Patrick J. Van Fleet · 2019
This chapter explains the wavelet packet transformation by looking at a schematic representation of an iterated wavelet transformation. It describes how the iterative transformation is computed. The large number of basis representations in a wavelet packet transformation is an advantage over an ordinary wavelet transformation since people have more options for decomposing the data. The chapter introduces the notion of a cost function that is used in conjunction with an algorithm for selecting the "best" representation. It uses the best basis algorithm in conjunction with the two-dimensional wavelet packet transformation to perform image compression. Probably the most notable application of wavelet packet transformation is the Federal Bureau of Investigation (FBI) Wavelet Scalar Quantization (WSQ) Specification developed by Bradley and Christopher Brislawn at Los Alamos National Laboratory and Hopper from the FBI. The chapter also discusses the WSQ method.