Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Null j2-stripes when Q2>1 in 1d ? #60

Open
lostanlen opened this issue Jan 24, 2014 · 0 comments
Open

Null j2-stripes when Q2>1 in 1d ? #60

lostanlen opened this issue Jan 24, 2014 · 0 comments
Labels

Comments

@lostanlen
Copy link
Member

Hello,
I don't understand the following behaviour of the 2nd-order scattering in 1d. Some coefficients are exactly zero when Q2>1 and j2<j1, even when path_margin == Inf. At first, I thought it was because the anti-aliasing filter (named lowpass) in conv_sub_1d doesn't intersect the range of values 'rng', but the same problem happens when filt_opt.format is 'fourier_multires' or 'fourier'.

Below is a minimal working example. Oddly enough, setting filt_opt.B = [8 1] seems to solve the problem ; why so ?

N = 32768;
load handel;
y = y(1:N);
T = 4096;
filt_opt.Q = [8 2];
%filt_opt.B = [8 1];
filt_opt.J = T_to_J(T,filt_opt);
filt_opt.format = 'fourier';
scat_opt.M = 2;
scat_opt.path_margin = Inf;
[Wop, filters] = wavelet_factory_1d(N, filt_opt, scat_opt);
[S,U] = scat(y, Wop);

set(figure(1),'WindowStyle','docked');
[~,sc2] = display_slice(S,1);
imagesc(log(abs(sc2)); % display energy in log-scale to see the black vertical stripes
figure(1);hold on;colormap copper;hold off;drawnow;commandwindow;

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

1 participant