| Re: problem with xroi continued.... [message #43454] |
Thu, 14 April 2005 13:49 |
Haje Korth
Messages: 651
Registered: May 1997
|
Senior Member |
|
|
Pravs,
I am afraid you are looking for functionality beyond what xroi can do for
you since it obviously does all evaluation on the byte scaled image, i.e.
you 0-2000 are squeezed into 0-255. You may just have to start doing some
programming here. The good thin is that idl provides all the commands you
need for you application.
Haje
"Pravs" <praveshsubramanian@yahoo.com> wrote in message
news:1113508669.308735.144030@l41g2000cwc.googlegroups.com.. .
> Hi Maarten,
>
> If it gets scaled to the 0-255 range, i will not get as accurate
> answers as I would otherwise. see, when you doing calculations, should
> not I have the original numbers than these scaled values?
>
> i tend to miss out on certain pixels because of that.
>
>
> pravs
> maarten wrote:
>> I am not sure if I understand your problem, but what is the problem
> with
>> bytscl?
>> It shouldn't matter whether the data is in the 0-2000 range, or the
>> 0-255 range, it is just a matter of scaling. You should probably not
>> specify a top and just let bytscl convert your full range of values
> to
>> the 0-255 range. At my computer this works perfectly well.
>>
>> cheers maarten
>>
>> Pravs wrote:
>>> Hi :
>>>
>>> Thanks for all the previous inputs. My guess is that since GE
> Genesis
>>> is a
>>> proprietary format, i may have to buy one of those tools available
> in
>>> the market such as DICOMatic. However, I have tried a crooked way
> to
>>> see if that helps.
>>>
>>> I converted the Genesis file to Analyze and from Analyze to DICOM
> using
>>> DICOMatic.
>>> I obtained a DICOM image too. However, the problem comes with XROI
> now.
>>> If I open the DICOM image directly from XROI, it is completely
> messed
>>> up. So, I try to open the image as xroi, bytscl(image).
>>>
>>> THIS DOES NOT SERVE THE PURPOSE AS THE SCALED IMAGE IS GRAYSCALE.
> EVEN
>>> WHEN I SPECIFY THE TOP, THE IMAGE HAS A MAX VALUE OF 255. THE
> ORIGINAL
>>> IMAGE HAS INTENSITY VALUES RANGING FROM 0-2000 BUT THE SCALED ONE
>>> BECOMES VERY LOSSY AND I LOSE PRECIOUS DATA.
>>> My question:
>>> What's the problem with XROI in opening such images that are in the
>>> proper format but not in 0-255 range? I guess if I cant get XROI to
>>> open it, I will make a program that can draw an ROI and I can
> further
>>> work with it
>>> ANy suggestions on that one?
>>>
>>> Thanks a lot, guys.
>>>
>>> Pravs
>>>
>
|
|
|
|
| Re: problem with xroi continued.... [message #43455 is a reply to message #43454] |
Thu, 14 April 2005 12:57  |
Pravesh
Messages: 17
Registered: August 2004
|
Junior Member |
|
|
Hi Maarten,
If it gets scaled to the 0-255 range, i will not get as accurate
answers as I would otherwise. see, when you doing calculations, should
not I have the original numbers than these scaled values?
i tend to miss out on certain pixels because of that.
pravs
maarten wrote:
> I am not sure if I understand your problem, but what is the problem
with
> bytscl?
> It shouldn't matter whether the data is in the 0-2000 range, or the
> 0-255 range, it is just a matter of scaling. You should probably not
> specify a top and just let bytscl convert your full range of values
to
> the 0-255 range. At my computer this works perfectly well.
>
> cheers maarten
>
> Pravs wrote:
>> Hi :
>>
>> Thanks for all the previous inputs. My guess is that since GE
Genesis
>> is a
>> proprietary format, i may have to buy one of those tools available
in
>> the market such as DICOMatic. However, I have tried a crooked way
to
>> see if that helps.
>>
>> I converted the Genesis file to Analyze and from Analyze to DICOM
using
>> DICOMatic.
>> I obtained a DICOM image too. However, the problem comes with XROI
now.
>> If I open the DICOM image directly from XROI, it is completely
messed
>> up. So, I try to open the image as xroi, bytscl(image).
>>
>> THIS DOES NOT SERVE THE PURPOSE AS THE SCALED IMAGE IS GRAYSCALE.
EVEN
>> WHEN I SPECIFY THE TOP, THE IMAGE HAS A MAX VALUE OF 255. THE
ORIGINAL
>> IMAGE HAS INTENSITY VALUES RANGING FROM 0-2000 BUT THE SCALED ONE
>> BECOMES VERY LOSSY AND I LOSE PRECIOUS DATA.
>> My question:
>> What's the problem with XROI in opening such images that are in the
>> proper format but not in 0-255 range? I guess if I cant get XROI to
>> open it, I will make a program that can draw an ROI and I can
further
>> work with it
>> ANy suggestions on that one?
>>
>> Thanks a lot, guys.
>>
>> Pravs
>>
|
|
|
|
| Re: problem with xroi continued.... [message #43471 is a reply to message #43455] |
Wed, 13 April 2005 07:13 |
maarten
Messages: 30
Registered: April 2002
|
Member |
|
|
I am not sure if I understand your problem, but what is the problem with
bytscl?
It shouldn't matter whether the data is in the 0-2000 range, or the
0-255 range, it is just a matter of scaling. You should probably not
specify a top and just let bytscl convert your full range of values to
the 0-255 range. At my computer this works perfectly well.
cheers maarten
Pravs wrote:
> Hi :
>
> Thanks for all the previous inputs. My guess is that since GE Genesis
> is a
> proprietary format, i may have to buy one of those tools available in
> the market such as DICOMatic. However, I have tried a crooked way to
> see if that helps.
>
> I converted the Genesis file to Analyze and from Analyze to DICOM using
> DICOMatic.
> I obtained a DICOM image too. However, the problem comes with XROI now.
> If I open the DICOM image directly from XROI, it is completely messed
> up. So, I try to open the image as xroi, bytscl(image).
>
> THIS DOES NOT SERVE THE PURPOSE AS THE SCALED IMAGE IS GRAYSCALE. EVEN
> WHEN I SPECIFY THE TOP, THE IMAGE HAS A MAX VALUE OF 255. THE ORIGINAL
> IMAGE HAS INTENSITY VALUES RANGING FROM 0-2000 BUT THE SCALED ONE
> BECOMES VERY LOSSY AND I LOSE PRECIOUS DATA.
> My question:
> What's the problem with XROI in opening such images that are in the
> proper format but not in 0-255 range? I guess if I cant get XROI to
> open it, I will make a program that can draw an ROI and I can further
> work with it
> ANy suggestions on that one?
>
> Thanks a lot, guys.
>
> Pravs
>
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
