Hey so I've written a snippet of code in order to perform an affine transformation and I am not getting the results I want. Can someone see what I am doig wrong.

Here I have an assignement to transform an image based on a user input. Below is what I have.

imshow(k)
p1 = impoint
Newp1 = impoint
p2 = impoint
Newp2 = impoint
p3 = impoint
Newp3 = impoint
p4 = impoint
Newp4 = impoint
p5 = impoint
Newp5 = impoint
POINT1= getPosition(p1)
POINT2= getPosition(p2)
POINT3= getPosition(p3)
POINT4= getPosition(p4)
POINT5= getPosition(p5)
NPOINT1= getPosition(Newp1)
NPOINT2= getPosition(Newp2)
NPOINT3= getPosition(Newp3)
NPOINT4= getPosition(Newp4)
NPOINT5= getPosition(Newp5)
X = [ POINT1 1 0 0 0; 0 0 0 POINT1 1; POINT2 1 0 0 0; 0 0 0 POINT2 1; POINT3 1 0 0 0; 0 0 0 POINT3 1; ...
    POINT4 1 0 0 0; 0 0 0 POINT4 1;  POINT5 1 0 0 0; 0 0 0 POINT5 1 ]
NPOINT1'
xprime = [NPOINT1';NPOINT2';NPOINT3'; NPOINT4';NPOINT5']
a = X \ xprime
A = [a(1:3)'; a(4:6)'; 0 0 1]
A = A;
A(7) = 0
A(8) = 0

above is a technique to find an affine transformation between two images. In this case though I am trying to use points on the same image in order to pull off this. I will demonstrate with an image below

: Do you know if this can actually work. I am using this program to use the affine transformation. Any feedback helps:

orange = [0 0 0]';
a = .15;
T = maketform('affine', A );
A2 = imread('latticenew.jpg');
h1 = figure; imshow(A2); title('Original Image')  
R = makeresampler({'cubic','nearest'},'fill');
B = imtransform(A2,T,R,'FillValues',orange);
h2 = figure; imshow(B);
title('Sheared Image');

The transformation is drastically different from what I want. And as you can see I have to set A(7) and A(8) to zero in order to use the maketform function although it outputs the parameter not close to zero.

Answer 1

[Walter Roberson]

I do not know anything about affine transforms. I only cleaned up the presentation of your question so that other people could read it more easily.