Figure changes during export

Hello everyone. I have created a loop where my function would save a plot of a series of posterior tangents overlaying vertebrae of the cervical spine. The problem is, after my function runs (code displayed below) only 5 out of the 6 lines appear on the plot. If I run it in isolation in the command window, all 6 lines appear. I have attached the images, the plot pre export with all 6 lines (ran in command window) and post export with the incorrect number (run as a function). I have gone through a number of times and cannot figure out why one line disappears when exporting. Any ideas?

Thanks in advance.

PS I have commented out parts of the code that are not relevant for this question. I did not delete so you were not confused about the input/output variables of my function.

function [c1_c7_angle,c2_c7_angle,c2c3_tangent,c3c4_tangent,c4c5_tangent,c5c6_tangent,c6c7_tangent]=rotation_angles(c1x,c1y,c2x,c2y,c3x,c3y,c4x,c4y,c5x,c5y,c6x,c6y,c7x,c7y,new_filename,i2)
%define posterior corners.
c2post=[c2x(1) c2y(1); c2x(4) c2y(4)];
c3post=[c3x(1) c3y(1); c3x(4) c3y(4)];
c4post=[c4x(1) c4y(1); c4x(4) c4y(4)];
c5post=[c5x(1) c5y(1); c5x(4) c5y(4)];
c6post=[c6x(1) c6y(1); c6x(4) c6y(4)];
c7post=[c7x(1) c7y(1); c7x(4) c7y(4)];

%c2_angle=rad2deg(atan((c2post(2,2)-c2post(1,2))/(c2post(2,1)-c2post(1,1))));
%c3_angle=rad2deg(atan((c3post(2,2)-c3post(1,2))/(c3post(2,1)-c3post(1,1))));
%c4_angle=rad2deg(atan((c4post(2,2)-c4post(1,2))/(c4post(2,1)-c4post(1,1))));
%c5_angle=rad2deg(atan((c5post(2,2)-c5post(1,2))/(c5post(2,1)-c5post(1,1))));
%c6_angle=rad2deg(atan((c6post(2,2)-c6post(1,2))/(c6post(2,1)-c6post(1,1))));
%c7_angle=rad2deg(atan((c7post(2,2)-c7post(1,2))/(c7post(2,1)-c7post(1,1))));

%c2c3_tangent=c2_angle-c3_angle;
%c3c4_tangent=c3_angle-c4_angle;
%c4c5_tangent=c4_angle-c5_angle;
%c5c6_tangent=c5_angle-c6_angle;
%c6c7_tangent=c6_angle-c7_angle;

%calculate c21-c7 angle by defining angle of inferior vertebral points.
%c1_angle=rad2deg(atan((c1y(1)-c1y(2))/(c1x(2)-c1x(1))));
%c2_inf_ang=rad2deg(atan((c2y(3)-c2y(4))/(c2x(3)-c2x(4))));
%c7_inf_ang=rad2deg(atan((c7y(3)-c7y(4))/(c7x(3)-c7x(4))));
%c1_c7_angle=c7_inf_ang-c1_angle;
%c2_c7_angle=c7_inf_ang-c2_inf_ang; 

vertebra_matrix=[c2post;c3post;c4post;c5post;c6post;c7post];
post_x=[c2x(1) c2x(4) c3x(1) c3x(4) c4x(1) c4x(4) c5x(1) c5x(4) c6x(1) c6x(4) c7x(1) c7x(4)];
post_y=[c2y(1) c2y(4) c3y(1) c3y(4) c4y(1) c4y(4) c5y(1) c5y(4) c6y(1) c6y(4) c7y(1) c7y(4)];
theta=zeros([1 6]);
x_diff=zeros([1 6]);
y_diff=zeros([1 6]);
V=zeros([6 2]);
factor_distance=2;

imshow(i2,'InitialMagnification',100);
hold on;

for h=1:6
    f=h*2;
    d=f-1;
    x_diff(h)=post_x(d)-post_x(f);
    y_diff(h)=post_y(f)-post_y(d);
    theta(h)=rad2deg(atan(y_diff/x_diff));
    V(h,:)=vertebra_matrix(d,:)-vertebra_matrix(f,:);
    pext(h,:)=vertebra_matrix(h,:)+V(h,:)*factor_distance;
    
    ax=gca;
    output_filename=sprintf('%s_rotation.tif',new_filename);
    exportgraphics(ax,output_filename);
    plot([post_x(f),pext(h,1)],[post_y(f),pext(h,2)],'r-')
end
end

Answer 1

[dpb]

    exportgraphics(ax,output_filename);
    plot([post_x(f),pext(h,1)],[post_y(f),pext(h,2)],'r-')

The exportgraphics call is before the call to plot() so the last one hasn't yet been plotted...when the loop finishes at the command line, then the line shows on the image--but that last image isn't exported.

Reverse the order of those two lines and joy should/will ensue...

[Nathan Hogaboom]

I figured it would be something silly like this. Newborn baby at home... my brain is mush.

Thanks.

[dpb]

Joy at home as well...congratulations! :)

You might go ahead and Accept the Answer so folks know is solved if nothing else...

[dpb]

OBTW, it would be more efficient to move the line

exportgraphics(ax,output_filename);

outside after the loop; no sense in rewriting the same file six times...one piece at a time; the final plot is the only one you're really interested in.

[Nathan Hogaboom]

Thank you!