imagesc problem

[febty febriani]

Hi, everyone.

I have a problem with imagesc command.

following is my script:

fh=fopen("fft.dat","r");
y=fscanf(fh,"%lf");
Fs=1;
step=1800;
window=4096;
[S, f, t] = specgram(y, 2*nextpow2(window), Fs, window, window-step);
S=S/max(max(S));
fclose(fh);
disp(size(S));imagesc(t,f,log((S)));set(gca,"ydir","normal");colorbar;axis([0 13530 0 0.1]);

the result of the first script is graph (attached).

I changed my script like below (please pay attention the red line (I added fliplr)). the result is graph1 (attached). 

fh=fopen("fft.dat","r");
y=fscanf(fh,"%lf");
Fs=1;
step=1800;
window=4096;
[S, f, t] = specgram(y, 2*nextpow2(window), Fs, window, window-step);
S=abs(S);
S=S/max(max(S));
fclose(fh);
disp(size(S));imagesc(t,f,log(fliplr(S)));set(gca,"ydir","normal");colorbar;axis([0 13530 0 0.1]);

I am just wondering why the dimension or segment (I have 8 segment on
my both result) on both result didn't change.
Logically,  because my
input data for spectrogram are not complete,  the first segment in
graph (the result of the first script) should be the last segment. Based on this assumption, I added fliplr to produce the second result (graph1), but dimension of the segments are still same.

Maybe, there are some  mistakes with my script,  any kind help is appreciated.
Thanks very much. 


-- 

******
febty febriani
Indonesian Institute of Sciences
Research Center for Physics
Kompleks PUSPIPTEK
Serpong, Indonesia, 15314

graph.png
Description: PNG image

graph1.png
Description: PNG image