Discussion :

[tubaas]

bonjour
j'ai une courbe de référence refCurve et une courbe mesurée lrMean.
la fonction ecart me permet de mesurer l'écart entre les deux.
j'aimerais translater la courbe de référence afin de minimiser cet écart, donc ajouter une valeur fixe à l'ensembles des points de la courbe mesurée.
pour ceci j'ai pensé à utiliser la fonction fminsearch mais elle me renvoie une erreur :

Code :

Sélectionner tout - Visualiser dans une fenêtre à part

fminsearch(@(y) ecart(refCurve,absc,lrMean+y,0),-10);

??? The following error occurred converting from struct to double:
Error using ==> double
Conversion to double from struct is not possible.

Error in ==> fminsearch at 205
fv(:,1) = funfcn(x,varargin{:});

merci si vous pouvez comprendre d'où peut venir le problème

[kmaniche]

Quel est le type et la taille de :

refCurve et lrMean

Assigning to an Empty Structure Field

Attempting to assign a structure to a field of another structure now results in an error if both of the following conditions are true:

• The field being assigned to has been initialized to an empty matrix.
• The field being assigned to is referenced in the assignment using an array index.

For example:

• mystruct.emptyfield = [];
  mystruct.emptyfield(1) = struct('f1', 25);
  ??? Conversion to double from struct is not possible.

This operation did not return an error in previous versions of MATLAB.

[Jean Dumoncel]

Salut,

Je pense que ton problème vient du 'lrMean+y' que tu passes en entrée de ta fonction. Essaie comme cela :

Code :

Sélectionner tout - Visualiser dans une fenêtre à part

fminsearch(@(y) ecart(y,refCurve,absc,lrMean,0),-10);

et modifie ta fonction ecart en conséquence.

[kmaniche]

When I run the following command:

property = blkproc(glcmPixel, [9 9], @(e) graycoprops(e());

I get the following error:

??? Conversion to double from struct is not possible.

Error in ==> blkproc at 150
b(brows, bcols) = firstBlock;

How can I fix it?

Thanks.

nrowb = size(glcmPixel,1) / 9;
ncolb = size(glcmPixel,2) / 9;
blocks = mat2cell(glcmPixel, repmat(9, 1, nrowb), repmat(9, 1, ncolb));
property = cellfun(@(e) graycoprops(e(), blocks);

> When I run the following command:

> property = blkproc(glcmPixel, [9 9], @(e) graycoprops(e());

graycoprops expects as input the N x N matrix (default 8 x 8)
that results from graycomatrix. You aren't giving it that
complete matrix: you are giving it the column-wise unraveling of
the matrix. The resulting statistics are likely to be meaningless.

Also, as the default output for graycomatrix is 8 x 8, it would
very likely be an error to be applying graycoprops to a 9 x 9
matrix -- unless you had specifically used 'NumLevels', 9
in the graycomatrix call.

My previous response in this topic copied your function behaviour
rather than giving what would be more likely the correct behaviour.
The code that would be more likely correct would be:

nrowb = size(glcmPixel,1) / 8;
ncolb = size(glcmPixel,2) / 8;
blocks = mat2cell(glcmPixel, repmat(8, 1, nrowb), repmat(8, 1, ncolb));
property = cellfun(@graycoprops, blocks);