about find_none_flectives_similarity

[lzg188]

thanks for your shared code, it is very good
but about this function named find_none_flectives_similarity i can not understand well.
like this code:
xy(Rect(1, 0, 1, xy.rows)).copyTo(A(Rect(0, xy.rows, 1, xy.rows)));//y
(xy(Rect(0, 0, 1, xy.rows))).copyTo(A(Rect(1, xy.rows, 1, xy.rows)));//-x
A(Rect(1, xy.rows, 1, xy.rows)) *= -1;
A(Rect(3, xy.rows, 1, xy.rows)).setTo(1.);
what are you going to do with adding the -x to matrix A, and why to inv the matrix trans_inv?
can you give me some explanation with that?
waiting for you replay!!!

[chinazy95]

1. what are you going to do with adding the -x to matrix A
   You should learn the meaning of "find_none_flectives_similarity". It means only considerate about rotation, scale(same as x and y), and translation.
   % Try varying these 4 parameters.
   scale = 1.2; % scale factor
   angle = 40*pi/180; % rotation angle
   tx = 0; % x translation
   ty = 0; % y translation

sc = scalecos(angle);
ss = scalesin(angle);

T = [ sc -ss 0;
ss sc 0;
tx ty 1];
so the transormed function worked:
X = ax+by+c
Y = -bx+ay+d

2. why to inv the matrix trans_inv?
   trans = find_similarity(src, dst);
   trans1 =find_none_flectives_similarity(uv, xy); //uv is src，xy is dst
   cv::solve(A, b, x, cv::DECOMP_SVD); //A is dst, b is uv
   so the result is dst to src. We are figuring out the src to dst, so it should be inverted.