[Insight-users] Re: some requested changes

[Luis Ibanez]

Hi Raghavendra,

The need for increasing the bound is not what
it is question, but rather the value of this
increase and the fact that the value is fixed
in the code.

I think Samson's suggestion of making this a
member variable that the user can set its the
way to go. We can still initialize this variable
to "0.001" if this value is a good default.

It is just not good in general to bury numerical
values in the code.

We could call it something like:

    m_BoundTolerance

or

    m_BoundIncrease


and provide Set/Get macros for getting access to it.

Would any of this names be ok with you ?
or would you prefere any more apropriate name ?


Thanks


   Luis


----------------------------------
Raghavendra Chandrashekara wrote:

> Hi Samson,
> 
> Thanks for your comments and finding the bugs in the test programs!
> 
> Luis Ibanez wrote:
> 
>> 3)  The min/max Fixed/Moving variables were added as you
>>     suggested.  I din't dare to remove the "evil" 0.001
>>     in this pass, so please keep this as an open issue
>>     we may have to pay a second visit to this file. 
> 
> 
> The reason that the upper bound was increased by a small amount was that 
> the itkHistogram class is designed not to hold a sample value equal to 
> the upper bound value of the histogram. I quote the message where there 
> was a discussion about this (05/13/2003) below. Since the images being 
> registered will contain these upper bound samples the upper bounds held 
> in the metrics are increased slightly. Maybe it would be better to 
> remove the 0.001 and let the user worry about setting the upper bound 
> appropriately for the images they are using? What do you think Luis?
> 
> Raghavendra
> 
>> Hi.
>>
>> The answer is no. It's by design. You have set the
>> upper bound a little bit larger than the largest value
>> you want to put into it. --- Raghavendra Chandrashekara 
>> <rc3 at doc . ic . ac . uk>
>> wrote:
>>  
>>
>>>> Hi Folks,
>>>>
>>>> Is it possible for the itk::Histogram class to hold
>>>> a sample value which is equal to the upper bound value? I've written a
>>>> simple test program which creates a 2D histogram and tries to increase
>>>> the frequency of a sample value at the lower and upper bounds. The
>>>> lower bound sample value frequency is increased by 1, but the upper 
>>>> bound is
>>>> not. Please can you tell me if this is a bug or this is the expected
>>>> behaviour.
>>>>
>>>> Thanks,
>>>>
>>>> Raghavendra
>>>
>>>   
>>>
>>>>> > #include "itkHistogram.h"
>>>>
>>>>     
>>>>
>>>> int main()
>>>> {
>>>>   typedef itk::Statistics::Histogram<double, 2>
>>>> HistogramType;
>>>>
>>>>   // Initialize histogram.
>>>>   HistogramType::Pointer pHistogram =
>>>> HistogramType::New();
>>>>   HistogramType::MeasurementVectorType lowerBound;
>>>>   HistogramType::MeasurementVectorType upperBound;
>>>>   HistogramType::SizeType size;
>>>>   size.Fill(1024);
>>>>   lowerBound.Fill(0);
>>>>   upperBound.Fill(255);
>>>>   pHistogram->Initialize(size, lowerBound,
>>>> upperBound);
>>>>
>>>>   // Add samples.
>>>>   HistogramType::MeasurementVectorType vec;
>>>>   vec[0] = 0;
>>>>   vec[1] = 0;
>>>>   pHistogram->IncreaseFrequency(vec, 1);
>>>>   vec[0] = 255;
>>>>   vec[1] = 255;
>>>>   pHistogram->IncreaseFrequency(vec, 1);
>>>>
>>>>   HistogramType::Iterator it = pHistogram->Begin();
>>>>   HistogramType::Iterator end = pHistogram->End();
>>>>   while (it != end) {
>>>>     if (it.GetFrequency() > 0) {
>>>>       std::cout << "Measurement = " <<
>>>> it.GetMeasurementVector()
>>>>         << "Frequency = " << it.GetFrequency() <<
>>>> std::endl;
>>>>     }
>>>>     ++it;
>>>>   }
>>>>
>>>>   return 0;
>>>> }
>>>
>>>   
>>>
>>>>> > PROJECT(Histogram)
>>>>
>>>>     
>>>>
>>>> INCLUDE(${CMAKE_ROOT}/Modules/FindITK.cmake)
>>>>
>>>> IF (ITK_USE_FILE)
>>>>   INCLUDE(${ITK_USE_FILE})
>>>> ENDIF (ITK_USE_FILE)
>>>>
>>>> INCLUDE_DIRECTORIES(${VIPITK_SOURCE_DIR}/Code)
>>>> LINK_DIRECTORIES(${VIPITK_BINARY_DIR}/Code)
>>>> LINK_LIBRARIES(${ITK_LIBRARIES})
>>>>
>>>> ADD_EXECUTABLE(Histogram Histogram.cpp)
>>>
>>>   
>>
>>
>>
>> =====
>> Jisung Kim
>> bahrahm at yahoo . com
>> 106 Mason Farm Rd.
>> 129 Radiology Research Lab., CB# 7515
>> Univ. of North Carolina at Chapel Hill
>> Chapel Hill, NC 27599-7515
>>
>> __________________________________
>> Do you Yahoo!?
>> The New Yahoo! Search - Faster. Easier. Bingo.
>> http://search . yahoo . com
>>  
>>
> 
> 
> _______________________________________________
> Insight-users mailing list
> Insight-users at itk . org
> http://www . itk . org/mailman/listinfo/insight-users
>