Wavelet Image Analyzer

Decompose and visualize images

Since R2023a

Description

The Wavelet Image Analyzer app enables you to visualize the discrete and continuous wavelet decomposition of images. With the Wavelet Image Analyzer app, you can:

Import images from your MATLAB^® workspace or from a file.
Vary wavelet transform parameters.
Easily compare different decompositions.
Reconstruct an image using discrete wavelet coefficient subbands you specify.
Export the image decompositions to your MATLAB workspace.
Generate MATLAB scripts to reproduce results in your workspace.

The Wavelet Image Analyzer app supports grayscale and RGB images.

Open the Wavelet Image Analyzer App

MATLAB Toolstrip: On the Apps tab, under Image Processing and Computer Vision, click the app icon.
MATLAB command prompt: Enter waveletImageAnalyzer.

Examples

Parameters

expand all

Discrete Wavelet Transform (DWT)

`Wavelet` — Wavelet family
`bior` (default) | `coif` | `db` | `fk` | `sym`

Wavelet family to use to obtain the decomposition, specified as:

Orthogonal Wavelet Families
- coif — Coiflets
- db — Daubechies extremal phase
- fk — Fejér-Korovkin
- sym — Symlets
Biorthogonal Wavelet Family
- bior — Biorthogonal spline

`Number` — Filter number
real number

Filter number for the selected wavelet family, specified as a real number. The filter number specifies which wavelet of the selected family to use to obtain the decomposition of the input image.

By default, the app uses the bior4.4 wavelet.

Example: To specify the Fejér-Korovkin wavelet with 14 coefficients, choose the fk wavelet family and the filter number 14.

`Level` — Wavelet decomposition level
`min(4,min(floor(log2(size(img,[1 2])))))`, where `img` is the input image (default) | positive integer

Wavelet decomposition level, specified as a positive integer.

Continuous Wavelet Transform (CWT)

`Wavelet` — Analyzing wavelet
`Morlet` (default) | `Cauchy` | `Endstop` | `Gaussian` | `Marr`

Wavelet to use in the 2-D CWT, specified as:

Isotropic Wavelets
- Gaussian — Gaussian
- Marr — Marr wavelet
Anisotropic Wavelets
- Cauchy — Cauchy wavelet
- Endstop — Endstop1 wavelet
- Morlet — Complex-valued Morlet wavelet

Anisotropic wavelets are sensitive to the orientation of features. Isotropic wavelets are not sensitive to orientations.

To see the equations that define the wavelets, use the cwtftinfo2 function.

Note

The Marr wavelet is real valued. If you use the Marr wavelet in the 2-D CWT of an image, the decomposition is also real valued.

`Normalization` — Normalization
`L1` (default) | `L2`

Normalization to use in the 2-D CWT, specified as:

L1 — The Fourier transform of the analyzing wavelet is multiplied by 1 at all scales.
L2 — The Fourier transform of the analyzing wavelet at a given scale is multiplied by the corresponding scale.

`Scale` — Scales
scalar | vector

Scales to use in the CWT, specified as a real-valued scalar or vector. Scales must be greater than or equal to 1. For more information, see CWT Scales in Wavelet Image Analyzer.

`Q-factor` — Number of voices per octave
`2` (default) | positive integer ≤ 4

Number of voices per octave to use in the CWT, specified as a positive integer between 1 and 4. The CWT scales are discretized using the specified number of voices per octave.

`Rotation` — Rotation angles
`4` (default) | positive integer ≤ 15 | vector

Rotation angles to use in the CWT, specified as a positive integer or vector. You can either specify an integer number of angles to evenly distribute in the interval [0, 2π), or choose specific angles in radians. By default, the angles are pi/2*[0:3]. For more information, see Viewing CWT Decomposition of Image in Wavelet Image Analyzer.

Programmatic Use

expand all