Error when performing imregtform with initial transformation
99 次查看(过去 30 天)
显示 更早的评论
I performed rigid 2d imregtform registration with initial transformation 'InitialTransformation'. It outputs the error:
'\mathworks\hub\3rdparty\R2022a\7296554\win64\ITK\include\ITK-5.2\itkRigid2DTransform.hxx:79:
Attempt to set a Non-Orthogonal matrix'
The rigid 2d registration is initialized with a transformation from point cloud ICP registration pcregistericp. pcregistericp outputs a rigidtform3d transform but with only 2d components as my point clouds are only 2d. I converted the point cloud rigidtform3d transformation back to rigidtform2d transformation.
Code is (data is attached):
% Pre-register with point cloud ICP registration
tform3d=pcregistericp(moving2dCloudPoint,fixed2dCloudPoint);
% Convert point cloud 3d transformation to 2d transformation
tforminit =rigidtform2d(tform3d.R(1:2,1:2),tform3d.Translation(1:2));
% Register with image registration intialized with 2d converted point cloud 3d transformation
tform2d=imregtform(fixedImage,R,movingImage,R,tformType,optimizer,metric,'InitialTransformation',tforminit);
Thanks for the help,
4 个评论
回答(1 个)
Gayathri
2024-11-18,6:24
I also faced the same issue at my end using the code and data provided. Please make sure that the rotation matrix in “tforminit” object must be an orthonormal matrix. I defined the “tforminit.R” to be a simple orthonormal matrix (which is also a rotation matrix), and the code ran without any errors as shown below.
load data
% Pre-register with point cloud ICP registration
tform3d=pcregistericp(moving2dCloudPoint,fixed2dCloudPoint);
% Convert point cloud 3d transformation to 2d transformation
tforminit =rigidtform2d(tform3d.R(1:2,1:2),tform3d.Translation(1:2));
tforminit.R=[0 -1;1 0];
% Register with image registration intialized with 2d converted point cloud 3d transformation
tform2d=imregtform(fixedImage,R,movingImage,R,tformType,optimizer,metric,'InitialTransformation',tforminit)
For more information on “imregtform” function, please refer to the following documentation link.
Hope you find this information helpful.
另请参阅
类别
在 Help Center 和 File Exchange 中查找有关 Point Cloud Processing 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!