Error using - Matrix dimensions must agree
1 visualizzazione (ultimi 30 giorni)
Mostra commenti meno recenti
simin as
il 22 Mag 2015
Commentato: simin as
il 23 Mag 2015
Hi .
Good time.
I have been trying to implement the AAM_2D_example model for prostat MRI images.The following error was encountered. Please help me. Error using - Matrix dimensions must agree.
Error in AAM_align_data2D (line 25) offsetr=-mean(rot-rotb);
Error in AAM_MakeShapeModel2D (line 19) [TrainingData(i).CVertices, TrainingData(i).tform]=AAM_align_data2D(TrainingData(i).Vertices,MeanVertices);
Error in AAM_2D_example (line 82) [ShapeData,TrainingData] = AAM_MakeShapeModel2D(TrainingData,options);
2 Commenti
dpb
il 22 Mag 2015
Well, you've given little really to go on other than
Error in AAM_align_data2D (line 25)
offsetr=-mean(rot-rotb);
is the line in question and so one must presume that
all(size(rot)==size(rotb))
returns FALSE To subtract arrays they have to have the same dimensions as the error message says, including the number of planes. Maybe one is color, the other not? Or, one is a subsection of another or whatever, but they clearly must not be conformant.
Walter Roberson
il 22 Mag 2015
This question appears to refer to this file exchange contribution: http://www.mathworks.com/matlabcentral/fileexchange/26706-active-shape-model--asm--and-active-appearance-model--aam-
Risposta accettata
Walter Roberson
il 22 Mag 2015
The subtraction will fail if the two sets of vertices being aligned do not have the same number of elements as built by LoadDataSetNiceContour.
I did not try to track down which files are the ones that differ, and I did not try to figure out how many vertices would be created. One thing I do see is that there must be exactly the same number of landmark points marked or else the number of vertices output would differ.
Più risposte (0)
Vedere anche
Categorie
Scopri di più su MRI in Help Center e File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!