Thread Subject: inverse continuous wavelet transform

Am I mistaken, or does the wavelet toolbox not contain a
function for the inverse continuous wavelet transform?

Basically, I'm looking for something like waverec, but for
continous wavelet transform (cwt). I looked through all the
documentation on the toolbox, searched the message board,
file exchange, but didn't see any.

Maybe I spazzed out and missed it. If it does not exist, I
am surprised mathworks didn't include it.

Thanks
jon

Jon,
  I didnt found the inverse Cont. WT at MAtlab.
  I suppose that such a tool is not implemented in the
comercial matlab licence, but we can found that in a
search at the internet.
Luiz

  


"jon erickson" <erickson.pleasenospam@caltech.edu> wrote
in message <fieqoj$4o9$1@fred.mathworks.com>...
> Am I mistaken, or does the wavelet toolbox not contain a
> function for the inverse continuous wavelet transform?
>
> Basically, I'm looking for something like waverec, but
for
> continous wavelet transform (cwt). I looked through all
the
> documentation on the toolbox, searched the message board,
> file exchange, but didn't see any.
>
> Maybe I spazzed out and missed it. If it does not
exist, I
> am surprised mathworks didn't include it.
>
> Thanks
> jon

For anyone who is interested: I have implemented a forward
and inverse CWT pair of functions. The fwd algorithm is
based off of code originally written by Torrence and Compo
at CU Boulder.

"Luiz Loures" <luiz_loures@hotmail.com> wrote in message
<fs337e$ego$1@fred.mathworks.com>...
> Jon,
> I didnt found the inverse Cont. WT at MAtlab.
> I suppose that such a tool is not implemented in the
> comercial matlab licence, but we can found that in a
> search at the internet.
> Luiz
>
>
>
>
> "jon erickson" <erickson.pleasenospam@caltech.edu> wrote
> in message <fieqoj$4o9$1@fred.mathworks.com>...
> > Am I mistaken, or does the wavelet toolbox not contain a
> > function for the inverse continuous wavelet transform?
> >
> > Basically, I'm looking for something like waverec, but
> for
> > continous wavelet transform (cwt). I looked through all
> the
> > documentation on the toolbox, searched the message board,
> > file exchange, but didn't see any.
> >
> > Maybe I spazzed out and missed it. If it does not
> exist, I
> > am surprised mathworks didn't include it.
> >
> > Thanks
> > jon
>

Dear Jon,

Could you please share your inverse continuous wavelet
transform algorithm? I tried one by myself, but
unfortunately it doesn't return the original signal accurately.

Kind regards,
Sebastian

"jon erickson" <erickson.pleasenospam@caltech.edu> wrote in
message <fsb4tn$1gi$1@fred.mathworks.com>...
> For anyone who is interested: I have implemented a forward
> and inverse CWT pair of functions. The fwd algorithm is
> based off of code originally written by Torrence and Compo
> at CU Boulder.
>

Wonderful man! There cannot be inverse continuous wavelet transform. Before trying to claim of developing the inverse continuous trasform have you ever imagined that computer is a discrete device?

On Apr 15, 7:10=A0pm, Yumnam Kirani Singh <kirani.si...@gmail.com>
wrote:
> Wonderful man! There cannot be inverse continuous wavelet transform. Befor=
e trying to claim of developing the inverse continuous trasform have you eve=
r imagined that computer is a discrete device?

Of course there can!!
It's only the mother wavelet that is continuous, i.e., she has a form
like DOG (derivative of Gaussian) or similar, as opposed to orthogonal
wavelet decomposition whose mother wavelet is discrete. The
implementation of the continuous wavelet transform of course MUST be
discrete.
Torrence & Compo wrote the bible on this. If you simply code up their
algorithm, you'll get the inverse. In fact, it's one line of code.

Jon,

It's wonderful you have implemented Inverse Continous
Wavelet Transform. As I'm working on a project where there
is a urgent need for it. could you please share the code
with me. I would feel highly obliged if you would mail me
the same personally at saurabh.nitj@gmail.com.

Thanking you in advance,

Saurabh

"jon erickson" <erickson.pleasenospam@caltech.edu> wrote in
message <fsb4tn$1gi$1@fred.mathworks.com>...
> For anyone who is interested: I have implemented a forward
> and inverse CWT pair of functions. The fwd algorithm is
> based off of code originally written by Torrence and Compo
> at CU Boulder.
>
> "Luiz Loures" <luiz_loures@hotmail.com> wrote in message
> <fs337e$ego$1@fred.mathworks.com>...
> > Jon,
> > I didnt found the inverse Cont. WT at MAtlab.
> > I suppose that such a tool is not implemented in the
> > comercial matlab licence, but we can found that in a
> > search at the internet.
> > Luiz
> >
> >
> >
> >
> > "jon erickson" <erickson.pleasenospam@caltech.edu> wrote
> > in message <fieqoj$4o9$1@fred.mathworks.com>...
> > > Am I mistaken, or does the wavelet toolbox not contain a
> > > function for the inverse continuous wavelet transform?
> > >
> > > Basically, I'm looking for something like waverec, but
> > for
> > > continous wavelet transform (cwt). I looked through all
> > the
> > > documentation on the toolbox, searched the message board,
> > > file exchange, but didn't see any.
> > >
> > > Maybe I spazzed out and missed it. If it does not
> > exist, I
> > > am surprised mathworks didn't include it.
> > >
> > > Thanks
> > > jon
> >
>

All, I will post the code on the mathworks website.
As NZTideMan points out, the inverseCWT code is (nearly) a
one line extension of Torrence and Compo's original code.

I've slightly modified what they posted on the web to
accommodate some other input/output options that I like, but
please note the heart of the original code is theirs, not mine.


"Saurabh Aggarwal" <saurabh.nitj@gmail.com> wrote in message
<g5tmtl$ffg$1@fred.mathworks.com>...
> Jon,
>
> It's wonderful you have implemented Inverse Continous
> Wavelet Transform. As I'm working on a project where there
> is a urgent need for it. could you please share the code
> with me. I would feel highly obliged if you would mail me
> the same personally at saurabh.nitj@gmail.com.
>
> Thanking you in advance,
>
> Saurabh
>
> "jon erickson" <erickson.pleasenospam@caltech.edu> wrote in
> message <fsb4tn$1gi$1@fred.mathworks.com>...
> > For anyone who is interested: I have implemented a forward
> > and inverse CWT pair of functions. The fwd algorithm is
> > based off of code originally written by Torrence and Compo
> > at CU Boulder.
> >
> > "Luiz Loures" <luiz_loures@hotmail.com> wrote in message
> > <fs337e$ego$1@fred.mathworks.com>...
> > > Jon,
> > > I didnt found the inverse Cont. WT at MAtlab.
> > > I suppose that such a tool is not implemented in the
> > > comercial matlab licence, but we can found that in a
> > > search at the internet.
> > > Luiz
> > >
> > >
> > >
> > >
> > > "jon erickson" <erickson.pleasenospam@caltech.edu> wrote
> > > in message <fieqoj$4o9$1@fred.mathworks.com>...
> > > > Am I mistaken, or does the wavelet toolbox not contain a
> > > > function for the inverse continuous wavelet transform?
> > > >
> > > > Basically, I'm looking for something like waverec, but
> > > for
> > > > continous wavelet transform (cwt). I looked through
all
> > > the
> > > > documentation on the toolbox, searched the message
board,
> > > > file exchange, but didn't see any.
> > > >
> > > > Maybe I spazzed out and missed it. If it does not
> > > exist, I
> > > > am surprised mathworks didn't include it.
> > > >
> > > > Thanks
> > > > jon
> > >
> >
>

p.s. by "post on the web" I mean that I will post in the
matlab file exchange.


"jon erickson" <erickson.pleasenospam@caltech.edu> wrote in
message <g62kqr$qs7$1@fred.mathworks.com>...
> All, I will post the code on the mathworks website.
> As NZTideMan points out, the inverseCWT code is (nearly) a
> one line extension of Torrence and Compo's original code.
>
> I've slightly modified what they posted on the web to
> accommodate some other input/output options that I like, but
> please note the heart of the original code is theirs, not
mine.
>
>
> "Saurabh Aggarwal" <saurabh.nitj@gmail.com> wrote in message
> <g5tmtl$ffg$1@fred.mathworks.com>...
> > Jon,
> >
> > It's wonderful you have implemented Inverse Continous
> > Wavelet Transform. As I'm working on a project where there
> > is a urgent need for it. could you please share the code
> > with me. I would feel highly obliged if you would mail me
> > the same personally at saurabh.nitj@gmail.com.
> >
> > Thanking you in advance,
> >
> > Saurabh
> >
> > "jon erickson" <erickson.pleasenospam@caltech.edu> wrote in
> > message <fsb4tn$1gi$1@fred.mathworks.com>...
> > > For anyone who is interested: I have implemented a forward
> > > and inverse CWT pair of functions. The fwd algorithm is
> > > based off of code originally written by Torrence and Compo
> > > at CU Boulder.
> > >
> > > "Luiz Loures" <luiz_loures@hotmail.com> wrote in message
> > > <fs337e$ego$1@fred.mathworks.com>...
> > > > Jon,
> > > > I didnt found the inverse Cont. WT at MAtlab.
> > > > I suppose that such a tool is not implemented in the
> > > > comercial matlab licence, but we can found that in a
> > > > search at the internet.
> > > > Luiz
> > > >
> > > >
> > > >
> > > >
> > > > "jon erickson" <erickson.pleasenospam@caltech.edu>
wrote
> > > > in message <fieqoj$4o9$1@fred.mathworks.com>...
> > > > > Am I mistaken, or does the wavelet toolbox not
contain a
> > > > > function for the inverse continuous wavelet transform?
> > > > >
> > > > > Basically, I'm looking for something like waverec,
but
> > > > for
> > > > > continous wavelet transform (cwt). I looked through
> all
> > > > the
> > > > > documentation on the toolbox, searched the message
> board,
> > > > > file exchange, but didn't see any.
> > > > >
> > > > > Maybe I spazzed out and missed it. If it does not
> > > > exist, I
> > > > > am surprised mathworks didn't include it.
> > > > >
> > > > > Thanks
> > > > > jon
> > > >
> > >
> >
>

Don't be ridiculous. There are a huge number of numerically
stable computer/discrete routines that approximate
continuous mathematics (integrals, for one).

Better to know first, and speak second.


Yumnam Kirani Singh <kirani.singh@gmail.com> wrote in
message
<21395913.1208243465919.JavaMail.jakarta@nitrogen.mathforum.org>...
> Wonderful man! There cannot be inverse continuous wavelet
transform. Before trying to claim of developing the inverse
continuous trasform have you ever imagined that computer is
a discrete device?