📄 mefig304.m
字号:
% mefig304 -- Segmented Wavelet Transform of "Cusp"; variable segmentations
%
% Here we compare the ordinary, non-segmented, wavelet transform
% of object "Cusp" with the segmented wavelet transform. We try
% segmentation points near to the actual discontinuity of the object.
% segmentation points closer to the actual discontinuity point give
% sparser looking wavelet transforms.
%
% Note: this script computes an object ent used by mefig305.
%
global Cusp n id
global E2 F2
global ent
%
L=5; D=2; t = id/n;
wc = FWT_AI(Cusp.^4,L,D,F2,E2);
%
%clf;
subplot(221)
PlotWaveCoeff(wc,L,300)
xlabel('t'); ylabel('dyad');
titl = sprintf('3.4a Ordinary Transform of Cusp^4');
title(titl)
plotnum =2;
%
eord = sum(sqrt(abs(wc(2^(L)+1:n))));
ent = zeros(4,16);
%
for spot = (id-8):(id+7),
tt = spot / n;
wc = FWT_SegAI(Cusp,L,D,F2,E2,tt);
ent(1,9+spot-id) = sum(sqrt(abs(wc(2^(L)+1:n))));
ent(2,9+spot-id) = sum((abs(wc(2^(L)+1:n))));
ent(3,9+spot-id) = - sum((wc(2^(L)+1:n).^2).* log(eps + wc(2^(L)+1:n).^2));
ent(4,9+spot-id) = sum(wc(2^(L)+1:n).^2);
if spot > (id-2) & spot < (id+2),
subplot(2,2,plotnum);
plotnum = plotnum+1;
PlotWaveCoeff(wc,L,300);
xlabel('t'); ylabel('dyad');
titl = sprintf('Transform Segmented at t=%g/2048',spot)
title(titl);
end
end
%
% Prepared for the paper Minimum Entropy Segmentation
% Copyright (c) 1994 David L. Donoho
%
%% Part of Wavelab Version 850% Built Tue Jan 3 13:20:41 EST 2006% This is Copyrighted Material% For Copying permissions see COPYING.m% Comments? e-mail wavelab@stat.stanford.edu
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -