vertical axis is flipped

Hi Matlab users: I am facing two strange problems. The first problem is with figure#1; it is inverted in y. The following code reads the .txt file ,which is attached, to be inverted in (y). I am wondering what could be wrong in my code that the data in the .txt file are inverted vertically. Second problem is with figure#4: Why the vertical axis is flipped in figure#4. The y axis shows the minus on top and the positive side on bottom. NOTE: this code produces the result in one minute, please run it and give me your comments.

clear all;
    close all;
    clc;
     x1=-2.5:1e-3:2.5;
            y1=-2.5:1e-3:2.5;

              step=30;   
              randomscannb=100;    
              lambda0=1.31;  
              rf=((2.25)/2); 
             [X1,Y1]=meshgrid(x1,y1);
             Ey2a4=exp((-X1.^2-Y1.^2)/rf^2);
              Ey1a4=dlmread('C:\Users\kgazz012\Desktop\March19th\ccccc_10t_10w_1nmstep.txt');
               Ey1a4c=conj(Ey1a4);
             sfinalmatrix=size(x1);sfinalmatrix=sfinalmatrix(2);
             figure(1);subplot(1,2,1);imagesc(real(Ey1a4));title('Ey1a4 real part')
                  axis square;
                  figure(2);subplot(1,2,1);imagesc([x1(1) x1(sfinalmatrix)],[y1(1) y1(sfinalmatrix)],real(Ey2a4));title('Ey2a4 real part');xlabel('um')
                  axis square;
                  figure(1);subplot(1,2,2);imagesc(imag(Ey1a4));title('Ey1a4 imag part')
                  axis square;
                  figure(2);subplot(1,2,2);imagesc([x1(1) x1(sfinalmatrix)],[y1(1) y1(sfinalmatrix)],imag(Ey2a4));title('Ey2a4 imag part');xlabel('um')
                  axis square;
                  fiberscan=Ey2a4+1;
                  selectedscan=round(rand(1,randomscannb)*(sfinalmatrix*sfinalmatrix)/(step^2));
                  match=0;
                  C=zeros(sfinalmatrix,sfinalmatrix);
              Norma = sum(sum(Ey2a4.*conj(Ey2a4)))*sum(sum(Ey1a4.*conj(Ey1a4)));
              c=0;
                tic
              for b=1:step:sfinalmatrix-300+1
                      for a=1:step:sfinalmatrix-248+1
                          c=c+1; 
                        display(strcat('fractionexecuted=',num2str(c*(step^2)/(sfinalmatrix*sfinalmatrix)))) 
                          fib=Ey2a4(b:b+300-1,a:a+248-1); 
                             fn=fib.*conj(Ey1a4); 
                          C(b+150,a+124)=sum(sum(fn))./sqrt(Norma); for k=1:randomscannb
                                      if c==selectedscan(k)
                                          match=1;
                                      end
                                  end

                                  if match==1
                                      fiberscan(b:b+300,a:a+5)=0;
                                      fiberscan(b:b+300,a+243:a+248)=0;
                                      fiberscan(b:b+5,a:a+248)=0;
                                      fiberscan(b+295:b+300,a:a+248)=0;
                                      match=0;
                                  end

                      end
                  end
                  toc
                  pack

                  result=abs(C);

          figure(100);imagesc([x1(1) x1(sfinalmatrix)],[y1(1) y1(sfinalmatrix)],real(fiberscan));title('Ey2a4 with 100 random scan positions indicated');xlabel('um')
      if step>1
                  resultx=real(result)*0;
                  for a=1:sfinalmatrix
                      for b=1:sfinalmatrix
                          if result(b,a)==0
                          elseif a>step+1 & b>step+1 & a<sfinalmatrix-step-1 & b<sfinalmatrix-step-1
                              hs=round(step/2);
                              resultx(b-hs:b+hs,a-hs:a+hs)=0*result(b-hs:b+hs,a-hs:a+hs)+result(b,a);
                          end
                      end
                  end
              else
                  resultx=(result);
              end

          figure(3);subplot(1,2,1);imagesc([x1(1) x1(sfinalmatrix)],[y1(1) y1(sfinalmatrix)],real(resultx));title('real part of overalp');xlabel('um')
          axis square;
          figure(3);subplot(1,2,2);imagesc([x1(1) x1(sfinalmatrix)],[y1(1) y1(sfinalmatrix)],imag(resultx));title('imaginary part of overlap');xlabel('um')
          axis square;

Answer 1

[Chad Greene]

Image axes are like that because image coordinates start at the top left corner if you're counting pixels. It's an old convention from image people, not from Matlab. Use

axis xy

[Naema]

Thanks Chad. Did you run the code?. I need to know the right axes because I need to know if my result(figure#4) is shifted up or down. I do not know how to use axis xy and where exactly in the code. Please, clarify.

[Chad Greene]

I did not run the code--it points to files that are not on my computer. Place axis xy right after any imagesc calls you want to flip. You can even combine it with your axis square lines like this

axis xy square

[Naema]

Why the field in the .txt file shows flipped vertically in figure #1. Does that means it is being used in the code while it is flipped?. Also, for figure #4, should I use the "Ticks" property to fix the flipped y-axis manually so that the minus side of "y-axis" is down and the positive side is up?

[Naema]

thansk, will try to use axis xy square. The txt file is attached in the question , do you see it?

[Chad Greene]

I don't have your text file. Can you upload an example image?

Messing with tick marks is probably not the best choice. It's usually better to plot the right data, than to apply manual fixes after plotting.

Try this:

load mandrill 
imagesc(X)

Then flip it like this:

axis xy

Flip it back:

axis ij

Or just plot it flipped from the beginning:

imagesc(flipud(X))

[Naema]

I already used : axis xy and the axis looks ok now but also the result was flipped. SO, is the flipped result the right one? Please run the txt file and fix it your way. I got confused:(

[Naema]

here is the .txt file attached

[Naema]

should not I only flip the y-axis but not the result? When I use axis xy, it flips the y axis and the plotted result.Please, let me now what is the correct thing to do as the plotted result is unknown to be shifted up or down.