Getting Error In FFT filtering ?

Question

jagannath mishra 2013-5-4

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/74569-getting-error-in-fft-filtering

sir i tried to make a gaussian filter.i use the steps given in the book(digital image processing using MATLAB,OF Gonzalez & Woods & Eddins )in the chapter3.i have done upto the step

    [f,revertclass]=tofloat(f)
here 'f' is an image(of int8),here also given an error that"Undefined function or method 'tofloat' for input arguments of type 'uint8'."
    PQ=paddedsize(size(f));
     F=fft2(f,PQ(1),PQ(2));
then i use gaussian filter and get the transfer function.and then
  H=ifftshift(H) for making center
then G=H.*F;
here i get an error which is "Error using ==> times Matrix dimensions must agree.
Error in ==> gaussian at 16
G=H.*F;"

please tell me how to overcome from this error

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Jordan Monthei 2013-5-8

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/74569-getting-error-in-fft-filtering#answer_84775

you are performing matrix multiplication when you use the '.*' command. The error you are receiving is telling you that the two "matrices" that you are trying to multiply are not of the same dimensions.

I assume you are trying to multiply numbers and not matrices and would recommend using '*' and not '.*'

3 个评论
显示 1更早的评论隐藏 1更早的评论

Image Analyst 2013-5-8

Jordan, what you said about multiplication is exactly the opposite of what it is. * is matrix multiplication and .* is element-by-element multiplication.

Jordan Monthei 2013-5-8