the problem of shifting and pcolor

Hi all, I am doing this code, and it appears that my text pixels are shifted compared to my pcolor output. Any suggestions. Note, my data contains NaN so image() does not work. Also, x and y are both of matrix types, not vectors. Any suggestions would be really appreciated. THANKS!
m_proj('stereographic','lat',90,'radius',60, 'rotateangle',270)
[lat_new,lon_new] = ndgrid(lat,lon);
[x,y] = m_ll2xy(lon_new,lat_new);
h = pcolor(x,y, cor); hold on;axis xy;
shading flat
sz = size(lat_new);
for i=1:sz(1)
for j=1:sz(2)
if prob(i,j)==1
%THESE APPEAR TO BE SHIFTED:(((
g = text(x(i,j),y(i,j),'*');
set(g,'fontsize',[10]);
set(g,'fontweight','bold');
hold on;
%z = text(x(i,j),y(i,j),num2str(cor(i,j)));
end
end
end

Antworten (2)

Image Analyst
Image Analyst am 4 Jul. 2012

0 Stimmen

Right. We talked about this before. Don't use pcolor. image() and imshow() have no problem handling images with nans in them - I actually tried and verified that. You just have to make sure x and y are in the range of 1 to sz(1) or sz(2)

4 Kommentare

Image Analyst
Image Analyst am 4 Jul. 2012
Bearbeitet: Image Analyst am 4 Jul. 2012
I'm not sure why you still insist on pcolor. Let me illustrate. Run this demo:
fontSize = 22;
m =[...
2 1 3 1 3
3 1 3 3 1
2 2 3 2 1
3 1 2 1 1
2 1 1 1 1]
figure(1);
pcolor(m)
shading('Faceted');
title('Faceted', 'FontSize', fontSize);
figure(2);
pcolor(m);
shading('interp');
title('interp', 'FontSize', fontSize);
figure(3);
pcolor(m);
shading('flat');
title('flat', 'FontSize', fontSize);
Do you expect to see a picture of 5 by 5 "pixels" where all the "1" elements have the same color, and all the "2" elements have the same color (but different than the color for 1), and all the "3" elements have the same color (but different than the color for 1 and 2)? That would seem reasonable wouldn't it? Is that how it actually looks? The answer is NO. The 1's DON'T all have the same color and neither do the 2's and 3's. Moreover the picture only has 4 by 4 pixels, not 5 by 5. You're warned about this in the source code for pcolor: " the last row and column of C are not used" Is that what you want? Plus there's the problem you originally asked. So, given all that, plus the fact that I told you image() and imshow() both handle (ignore) nans, can you explain to me why you still want to use pcolor?
jenka
jenka am 4 Jul. 2012
Hi, I am trying to use image(). However, it gives me the entire screen as blue:( Any suggestions? m_proj('stereographic','lat',90,'radius',60, 'rotateangle',270) [x,y] = m_ll2xy(lon_new,lat_new); %note here I use lon and lat (vectors) h = image(lon,lat, cor) sz = size(x); for i=1:sz(1) for j=1:sz(2) if prob(i,j)==1 %THESE APPEAR TO BE SHIFTED:((( g = text(x(i,j),y(i,j),'*'); set(g,'fontsize',[10]); set(g,'fontweight','bold'); hold on; end end end
sorry here is the code use:
m_proj('stereographic','lat',90,'radius',60, 'rotateangle',270)
[lat_new,lon_new] = ndgrid(lat,lon);
[x,y] = m_ll2xy(lon_new,lat_new);
%lon and lat are vectors
h = image(lon,lat, cor);
sz = size(lat_new);
for i=1:sz(1)
for j=1:sz(2)
if prob(i,j)==1
%THESE APPEAR TO BE SHIFTED:(((
g = text(x(i,j),y(i,j),'*');
set(g,'fontsize',[10]);
set(g,'fontweight','bold');
hold on;
end
end
end
I don't have those functions. If they're in the mapping toolbox, I don't have that. You can use a colormap to get rid of the blue, like
imshow(yourArray, []);
colormap(gray(256));

Melden Sie sich an, um zu kommentieren.

jenka
jenka am 4 Jul. 2012

0 Stimmen

This is exactly the issue with pcolor I am facing. I am just not sure how to solve this small problem: p_color with shading flat will draw a panel between the (i,j),(i+1,j),(i+1,j+1),(i,j+1) coordinates of the X/Y matrices with a color corresponding to the data value at (i,j). Thus everything will appear shifted by one half a pixel spacing.

2 Kommentare

Walter Roberson
Walter Roberson am 4 Jul. 2012
Yes, that is how it is designed, and is the reason not to use pcolor()
Chad Greene
Chad Greene am 1 Mai 2015
Here's a depiction of some funny effects using pcolor.

Melden Sie sich an, um zu kommentieren.

Gefragt:

am 4 Jul. 2012

Kommentiert:

am 1 Mai 2015

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by