dwt3d instruction returns with error saying filter H must be a double precision filter

when i try to run the following example code for dwt3D

    [af, sf] = farras;
    x = rand(128,64,64);
    w = dwt3D(x,3,af);
    y = idwt3D(w,3,sf);
    err = x-y; 
    max(max(max(abs(err))))

i get an error saying

Error using upfirdn>validateinput (line 104)
The filter H must be a double-precision vector.

Error in upfirdn (line 81)
[p,q] = validateinput(x,h,varargin);

Error in afb3D_A (line 41)
   lo(:, :, k) = upfirdn(x(:, :, k), lpf, 1, 2);

Error in afb3D (line 36)
[L, H] = afb3D_A(x, af1, 1);

Error in dwt3D (line 27)
    [x w{k}] = afb3D(x, af, af, af);

i have all the required function files with me in the current folder of MATLAB

[Walter Roberson]

This appears to relate to http://eeweb.poly.edu/iselesni/WaveletSoftware/standard3D.html .

[Amit Singh]

yes tried that but it did not work. adding a image of the error

[Walter Roberson]

That code works for me in R2018a.

[Amit Singh]

can you attach the dwt3D.m and afb3D.m files?

[Walter Roberson]

http://eeweb.poly.edu/iselesni/WaveletSoftware/allcode.zip

[Amit Singh]

thank you have been trying using the same function files but doesn't run on my matlab R2017b or R2013a

[Walter Roberson]

I just tested in both R2017b and R2013a, and the code works fine for me.

After you execute that first line of code,

    [af, sf] = farras;

what shows up for

 class(af)

? It should show up as double(), and if it does not then you are not using the farras.m that is provided with the code.

[Amit Singh]

when i execute class(af) i get it as 'cell' not double

[Walter Roberson]

What shows up for

which farras
which dwt3D

[Amit Singh]

thank you its working now :)

Answer 1

[Walter Roberson]

Conflicting versions of the data loading function caused this error. The required data is in the form of numeric arrays; the data loading function was returning a cell array instead.