[Insight-users] 3D Lidar-point-to-Image registration using pointSetToImageRetgistration

[Luis Ibanez]

Hi Ruisheng,

Unfortunately we don't have specific examples for this
configuration of registration (Lidar-to-Image) in the
ITK source tree.

However,

You may find useful to look at the code of the test:

   Insight/Testing/Code/Algorithms/
         itkPointSetToImageRegistrationTest_1.cxx


   Regards,


          Luis

---------------------------------------------------------------
On Mon, Apr 12, 2010 at 3:51 PM, ruisheng wang <wangruisheng at gmail.com>wrote:

> Hi Luis,
>
> Thank you very much for your replies. Do you have an example code showing
> the process of Lidar-to-Image registration using
> Mutual informaiton. I could not find examples for doing this. In the ITK
> document, there is not an example showing
> PointSetToimageRegistration, do you have an example code showing how to use
> PointSetToImageregistration?
>
> thanks a lot!
> Ruisheng
>
>
>
> Hi Ruisheng,
>
> In order to use this class effectively, you should
> first become very familiar with the ITK Registration
> Framework.
>
> Please read the ITK Software Guide:
>
>       http://www.itk.org/ItkSoftwareGuide.pdf
>
> In particular the Chapter 8 "Image Registration"
> in pdf-pages:  347-533.
>
>
> Once you are familiar with the framework,
> you may want use the following configuration:
>
>
> 1) Use the PointSetToImageRegistration class
>
> 2) Use the itkPointSetToImageMetric class
>
> 3) Use the perspective transform
>
> 4) Start with an optimizer that does not require
>    derivatives (e.g. the Amoeba optimizer).
>
>
> Key factors for a successful registration are:
>
> A)  Good initialization of the Transform
>
> B)  Steady setup for the optimizer parameters
>
> C) Proper scaling of the Transform parameters.
>
> In your case, please note that (A) includes to
> make a very good initialization for the center
> of projection and the focal distance, and that
> (B) means that you should start with optimizer
> settings that barely modify the Transform, and
> from them you can progress towards having
> the Transform explore a wider range of
> parameters.
>
>
>      Regards,
>
>
>             Luis
>
>
> Hi everyone,
>
> I'm trying to register 3D lidar with 2D optical images using the transform
> ITK::
> Rigid3DPerspectiveTransform.
> But seems there is not much explanation or examples from ITK to describe
> the this functions. I wonder if
> anyone here could give me some hints for how to use
> ITK::Rigid3DPerspectiveTransform for the 3D-to2D registration
> problems.
>
> Thanks very much!
> Ruisheng
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20100415/373124ae/attachment.htm>