If the above application does not start, please install OpenJDK from adoptium.net.
This applet applies the wavelet decomposition to the 2-d images.
You can select the basis of the wavelet between the Haar and the Daubecies (N=3).
(Note that the decomposition with the Daubecies basis takes more time than that of Haar.)
When the "wavelet" button is pressed,
the image is decomposed into the four groups of components, namely,
low-frequency components and
high-frequency components in the y-direction, in the x-direction,
and in the xy-direction.
Those components are aligned in the right panel of the applet
in the following order.
low-freq. components
high freq. components in y-direction
high freq. components in x-direction
high freq. components in xy-direction
By applying the decomposition to the low-frequency components again,
the four components of the image of lower resolution
are obtained.
These components with lower resolutions are obtained
by pressing the "wavelet" button successively.
The image recognition using the wavelet components
with various resolutions is called multiresolution analysis.