matlab code
sabeena naushad
only supports ROI coding.
.
I' ve generated the ROI mask by 'roipoly' function in matlab.
Find out integer wavelet transform using lifting schemes (In
the paper they insisted to do so for perfect reconstruction)
Quantised background coefficients using
equation Q(X) = sign(X)*floor((abs(X)./qi)) where qi is quantisation
index which I selected at random.
ROI coefficients are not quantised.
Next. bitplane conversion of the coefficients has to be done.Then
scaling of ROI bits withrespect to BG.Then comes entropy coding.
In the paper the authors claim about reconstruction
quality.To get final result decoding also has to be done.
Entropy encoding is based on QM coder of arithmetic coder.
For bit plane conversion is there any specific function in matlab. If anyone has matlab code for JPEG2000 please send me through my email
It will be very helpful if you could help me.
regards
sabeena
Walter Roberson
> If anyone has matlab code for JPEG2000 please send me through my email
> It will be very helpful if you could help me.
JPEG2000 is subject to several patents. Some of the patent holders have waved
royalty fees for use of the code with JPEG2000, but other patent holders
have not. I have not yet found a list of which patents have their royalties
waved and which ones require that royalties be paid under what circumstances.
Historically, it has not been uncommon for -decoding- to be royalty free
in compression programs, but that fees have had to be paid for -encoding-
programs. *Including*, in some cases, student programs.
I would thus not expect that anyone would send you JPEG2000 code in Matlab,
due to the patent issues.
ImageAnalyst
> > It will be very helpful if you could help me.
>
> royalty fees for use of the code withJPEG2000, but other patent holders
> have not. I have not yet found a list of which patents have their royalties
> waved and which ones require that royalties be paid under what circumstances.
>
> Historically, it has not been uncommon for -decoding- to be royalty free
> in compression programs, but that fees have had to be paid for -encoding-
> programs. *Including*, in some cases, student programs.
>
> due to the patent issues.
---------------------------------------------------------------------------------------------------
jpeg2000 is an ISO standard, like PDF. They are trying to avoid any
patent squabbles by changing the standard to avoid patented algorithms
or get the inventors (11 of them so far) to give license free usage.
It's a murky situation. The whole issue is explained (still somewhat
confusingly) in the FAQ link on this page:
http://www.jpeg.org/jpeg2000/
According to the FAQ, the current version of jpeg2000 is fee free (for
those patents that it includes), and has avoided use of other
patents. Bottom line is that I think you're "free" to use it. At
least that's how I read it.
Walter Roberson
> jpeg2000 is an ISO standard, like PDF. They are trying to avoid any
> patent squabbles by changing the standard to avoid patented algorithms
> or get the inventors (11 of them so far) to give license free usage.
> It's a murky situation. The whole issue is explained (still somewhat
> confusingly) in the FAQ link on this page:
> http://www.jpeg.org/jpeg2000/
> According to the FAQ, the current version of jpeg2000 is fee free (for
> those patents that it includes), and has avoided use of other
> patents. Bottom line is that I think you're "free" to use it. At
> least that's how I read it.
The FAQ says that it is *intended* as royalty and license-fee free, but not patent-free,
and the FAQ specifically says, http://www.jpeg.org/jpeg2000/j2kpart1.html
>>Part 1 also includes guidelines and examples, a bibliography of technical references, and a list of >>companies from whom patent statements have been received by ISO. JPEG 2000 was developed with the >>intention that Part 1 could be implemented without the payment of licence fees or royalties, and a >>number of patent holders have waived their rights toward this end. However, the JPEG committee
>> cannot make a formal guarantee, and it remains the responsibility of the implementer to ensure
>> that no patents are infringed.
I can't afford to order the official ISO standard (it costs real money), but the last
draft http://www.jpeg.org/jpeg2000/CDs15444.html in Annex L lists 22 companies or
organizations which have filed claims that a patent of theirs is being used by JPEG2000.
There is nothing in that document that indicates which of the patent holders have waived
royalties or under which circumstances, and the introduction to Annex L {draft} reads:
>>>There is the possibility that, for some of the processes specified in this
>>>Recommendation | International Standard, conformance or compliance may require
>>>use of an invention covered by patent rights.
>>>By publication of this Recommendation | International Standard, no position is
>>>taken with respect to the validity of this claim or of any patent rights in
>>>connection therewith. Information regarding such patents can be obtained from
>>>the any organizations. The table summarizes the formal patent and intellectual
>>>property rights statements that have been received.
What does this mean? It means that before starting work on any JPEG2000 code, one
must obtain the official standard (which presumably has the latest list of claimants)
and one must then go through and contact EACH of those 22 or more organizations and
ask them which patents they are asserting with respect to this standard and ask them
for formal copies of their license terms and associated fees. And hope that you
are -able- to track them down, since contact information is not given, and hope that
they bother to answer you... that's a minimum of 22 different organizations that you
have to check with before you write *any* JPEG2000 code, just to be sure that you
are legally safe with the -known- patent claims!
If it were me, I wouldn't touch the situation except to hand it over to our corporate
lawyers to do the legal analysis and chasing -- and the first thing they would say
to me, I'm sure, is "Considering the number of known patent encumbrances involved,
are you *certain* that your work absolutely -requires- you to implement this,
and *certain* that it is important enough to your work to make it worth the several
thousand dollars it will take for our legal staff to get the clearances???"
