[Insight-users] Problem with 2D Registration (local minima)

Thu Oct 22 11:52:23 EDT 2009

The optimizer is stopping after reaching the maximum number of
iterations. This is normally not the best way to terminate an
optimization.

What do you expect for the optimum value of the parameters?

Perhaps you need a larger maximum step length? Or perhaps, as Luis
suggested, the starting point is too far from the optimum for this
optimizer.

Bill

On Thu, Oct 22, 2009 at 10:09 AM, Manuel David Tabas Calle
<manueldavid.tabas at hotmail.es> wrote:
>
> Hi Luis,
>
> Thank you for answering me.
>
> 1) As you said, my images are from Microscopy and the spacing is about 2.85
> micrometres/pixel. The images are in a raw format and I used the following
> .mhd file for reading it:
>
>
>     NDims = 2
>     DimSize = 2000 2000
>     ElementType = MET_UCHAR
>     ElementSpacing = 1.0 1.0
>     Offset = 0.0 0.0
>     ElementByteOrderMSB = False
>     ElementDataFile = fixed.raw
>
> I also tried to as ElementSpacing 2.83 and 0.00283 (just in case...) but the
> results weren´t good either.
>
> 2) I also try to put a relaxation factor of 0.9 as you said and I couldn´t
> see any improvement.
>
> 3)  I´m attaching three .txt files with some results obtained.
> The results1.txt file shows the output of the program as you can find it on
> ITK/Examples/Registration/.
> The results2.txt file shows the output of the program with a predifined
> center in the position 1000x1000. The results3.txt file shows the output of
> the program with a predifined center in the position 1000x1000, a predefined
> angle of 0.9 radians and a predefined traslation of -340 in the X axis and
> 800 in the Y axis (that are the values for an optimal registration).
>
>
> I hope you can help me to fix my problem with that information... an be free
> to ask anything else :P.
>
> Thank you really much.
>
> Manuel.
>
>
>> Date: Wed, 21 Oct 2009 17:29:08 -0400
>> Subject: Re: [Insight-users] Problem with 2D Registration (local minima)
>> From: luis.ibanez at kitware.com
>> To: manueldavid.tabas at hotmail.es
>> CC: insight-users at itk.org
>>
>> Hi Manuel,
>>
>> Thanks for the detailed description of your problem.
>>
>> A couple of questions:
>>
>>
>> 1) What is the pixel spacing of your image ?
>>
>> The typical proportion of :
>>
>> > const double translationScale = 1.0 / 1000.0;
>>
>> is set up for images whose physical extent in
>> the units used by the spacing is in the range
>> of hundreds.
>>
>> For example, a typical CT scan will have 512x512
>> pixels with a spacing close to 1mm / pixel.
>>
>> Your images are probably from Microscopy or
>> from a Satellite, so I'm wondering about the units
>> that you use (microns or kilometers) and the
>> exact numeric values of the spacing.
>>
>>
>> 2) Could you post to the list the output of the
>> command observer that tracks the values of
>> the metric and the transform at every iteration ?
>>
>>
>> Something that happens often when using the
>> RegularStepGradientDescentOptimizer is that,
>> if the registration process have a couple of
>> bounces of direction in the parametric space,
>> the step length may be reduced too fast, and
>> not give the optimizer the oportunity to walk
>> far enough to get to the solution.
>>
>>
>> You may want to try using the method:
>>
>> optimizer->SetRelaxationFactor( 0.9 );
>>
>>
>> This method defines the factor by which the
>> step length is reduced at every time that the
>> parametric space step changes direction.
>> This may prevent the premature shrinkage
>> of the step length (assuming that this is the
>> real problem).
>>
>>
>> Anothe typical issue: It may simply be that you
>> need a better initialization for your images.
>>
>>
>> Note that the Transform initializer in the example:
>>
>> ImageRegistration6.cxx
>>
>> is using the methods MomentsOn().
>>
>> This may or may not be the right method to use,
>> depending on the content of your images.
>>
>> For example: If the two images have different
>> modalities, or if the main features of one image
>> may not overlap to the one in the other image.
>>
>> If you could post a screenshot of both images
>> we could avoid all this speculation... :-)
>>
>> Or...
>> If you can share your images in a public web
>> site, that will be even better.
>>
>>
>>
>> Thanks
>>
>>
>> Luis
>>
>>
>> ----------------------------------------------------------------
>> On Wed, Oct 21, 2009 at 11:56 AM, Manuel David Tabas Calle
>> <manueldavid.tabas at hotmail.es> wrote:
>> > Hello ITK,
>> >
>> > I´m working with ITK to do registration of 2D images and I think I reach
>> > a
>> > local minima that makes a poor registration.
>> >
>> > I´m using 2D images 2000x2000 pixels in what the moving image needs a
>> > rotation and a translation. So I reproduced the ImageRegistration6.cxx
>> > to do
>> > that. But the result I got is not good enough so I think that the
>> > MeanSquares metric reaches a local minima. The value of the metric is
>> > about
>> > 4000 (aprox.), and I run a test using as input parameters the optimal
>> > ones
>> > (that is, the result I think i should obtain) and the metric returns a
>> > value
>> > of 2500 (aprox.).
>> >
>> > The framework I use is:
>> >
>> >    itkImageRegistrationMethod
>> >    itkMeanSquaresImageToImageMetric
>> >    itkLinearInterpolateImageFunction
>> >    itkRegularStepGradientDescentOptimizer
>> >    itkCenteredRigid2DTransform
>> >
>> > And the parameters of my registration are:
>> >
>> > const double translationScale = 1.0 / 1000.0;
>> > optimizerScales[0] = 1.0;
>> > optimizerScales[1] = translationScale;
>> > optimizerScales[2] = translationScale;
>> > optimizerScales[3] = translationScale;
>> > optimizerScales[4] = translationScale;
>> >
>> > optimizer->SetMaximumStepLength( 0.1 );
>> > optimizer->SetMinimumStepLength( 0.001 );
>> > optimizer->SetNumberOfIterations( 200 );
>> >
>> >
>> > The scales are well defined and I tried to vary the parameters of the
>> > optimizer but the results were similar, so I don´t know what to do next.
>> > Any
>> > idea??
>> >
>> > Thank you in advance, I hope you can have more ideas to try.
>> >
>> >
>> >
>> > Manuel.
>> >
>> >
>> >
>> >
>> > _____________________________________
>> > Powered by www.kitware.com
>> >
>> > Visit other Kitware open-source projects at
>> > http://www.kitware.com/opensource/opensource.html
>> >
>> > Please keep messages on-topic and check the ITK FAQ at:
>> > http://www.itk.org/Wiki/ITK_FAQ
>> >
>> > Follow this link to subscribe/unsubscribe:
>> > http://www.itk.org/mailman/listinfo/insight-users
>> >
>> >
>
> ________________________________
> Vive toda la experiencia de Messenger con Orange desde tu Blackberry
> ¡Descúbrelo!
> _____________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the ITK FAQ at:
> http://www.itk.org/Wiki/ITK_FAQ
>
> Follow this link to subscribe/unsubscribe:
> http://www.itk.org/mailman/listinfo/insight-users
>
>