wptree

WPTREE constructor

Syntax

T = wptree(order,depth,x,wname)

T = wptree(order,depth,x,wname,enttype,entpar)

T = wptree(order,depth,x,wname,enttype,entpar,userdata)

Description

T = wptree(order,depth,x,wname) returns a complete wavelet packet tree T of order order corresponding to a wavelet packet decomposition of x at level depth, using Shannon entropy and the wavelet specified by wname.

T = wptree(order,depth,x,wname) is equivalent to T = wptree(order,depth,x,wname,'shannon').

T = wptree(order,depth,x,wname,enttype,entpar) uses the entropy type specified by enttype. entpar is an optional parameter depending on the value of enttype.

T = wptree(order,depth,x,wname,enttype,entpar,userdata) sets the userdata field of T.

Examples

Create Wavelet Packet Tree

Create a 1-D signal.

x = rand(1,512);

Create the wavelet packet decomposition tree associated with the wavelet packet decomposition of the signal at level 3 using the db3 wavelet.

t = wptree(2,3,x,"db3");

Recompose the fourth and fifth nodes of the tree. Plot the result.

t = wpjoin(t,[4;5]);
plot(t)

Click the node (3,0) to get this figure:

Input Arguments

order — Order of the tree

2 | 4

Order of the tree, specified as 2 or 4. The order of the tree is the number of children of each nonterminal node. If x is a vector (1-D signal), specify an order of 2. If x is a matrix (image), specify an order of 4.

Data Types: double

depth — Level of wavelet packet decomposition

positive integer

Level of wavelet packet decomposition, specified as a positive integer.

Data Types: double

x — Input data

vector | matrix

Input data, specified as a vector (signal) or matrix (image).

Data Types: double

wname — Wavelet name

character vector | string scalar

Wavelet name, specified as a character vector or string scalar. For more information, see wfilters.

Data Types: string | char

enttype — Entropy type

character vector | string scalar

Entropy type, specified as a character vector or string scalar. For more information, see wpdec or wpdec2

Data Types: string | char

entpar — Optional parameter

real number | character vector | string scalar

Optional parameter used for entropy computation. For more information, see wpdec or wpdec2.

Data Types: double | string | char

userdata — User data

array | cell array | structure array

User data to set in the userdata field of T, specified as an array, cell array, or structure array.

Example: t = wptree(2,3,x,'db3','sure',0.5,{1,"aa",rand(3,3)})

Output Arguments

T — Wavelet packet tree

WPTREE object

Wavelet packet tree, returned as a WPTREE object.

• If order = 2, T is a WPTREE object corresponding to a wavelet packet decomposition of the vector (signal) x, at level depth with a particular wavelet wname.

• If order = 4, T is a WPTREE object corresponding to a wavelet packet decomposition of the matrix (image) x, at level depth with a particular wavelet wname.

The WPTREE object has these fields:

'dtree'	DTREE parent object
'wavInfo'	Structure (wavelet information)
'entInfo'	Structure (entropy information)

For more information on object fields, see the get function or type:

help wptree/get

The wavelet information structure, 'wavInfo', contains

'wavName'	Wavelet name
'Lo_D'	Low Decomposition filter
'Hi_D'	High Decomposition filter
'Lo_R'	Low Reconstruction filter
'Hi_R'	High Reconstruction filter

The entropy information structure, 'entInfo', contains

'entName'	Entropy name
'entPar'	Entropy parameter

Fields from the DTREE parent object:

'allNI'

All nodes information

'allNI' is an array of size nbnode by 5, which contains

ind	Index
size	Size of data
ent	Entropy
ento	Optimal entropy

Each line is built based on this scheme:

Version History

Introduced before R2006a