thanks DK... what if its a perspective transformation and not affine exactly? cos i tried using this matrix [xnew ynew 1]=[a b c;d e f;0 0 1][x y 1]
now after getting the matrix parameters using the feature points matched in both the images , when i put the border pixels as input to this matrix the transformed pixels do not give the border exactly in the new image contaning this object On Tue, Jun 21, 2011 at 6:14 PM, DK <[email protected]> wrote: > If the two preconditions for an affine relationship are met: > The collinearity <http://en.wikipedia.org/wiki/Line_(geometry)> relation > between points; i.e., the points which lie on a line continue to be > collinear after the transformation > Ratios of distances along a line; i.e., for distinct collinear points *p*1 > ,*p*2,*p*3, the ratio | *p*2 - *p*1 | / | *p*3 - *p*2 | is preserved > > Then solve the affine transformation relationships: > > if X is the source image feature point locations and Y is the target image > feature point locations and they are related by > an affine transform (rotations + transformations) then you can create an > affine transformation equation. > > Y = A X + b > or > Y = A' X where A' is the homogeneous representation of the Affine > transform. > > Solve this set of equations and ensure consistency (no. of variables < no. > of equations). > > If you don't know which feature point maps to which other, then solve the > equations for > all possible combinations and see which one fits the best. (You'll have to > be slightly lax > about the consistency constraints). > > You may want to see the LUP algorithm for solving equations in matrix form. > Also see: http://en.wikipedia.org/wiki/Affine_transformation (section: > affine transformations of a plane) > > -- > DK > > http://twitter.com/divyekapoor > http://www.divye.in > > -- > You received this message because you are subscribed to the Google Groups > "Algorithm Geeks" group. > To view this discussion on the web visit > https://groups.google.com/d/msg/algogeeks/-/z9x9sVs3eCoJ. > > To post to this group, send email to [email protected]. > To unsubscribe from this group, send email to > [email protected]. > For more options, visit this group at > http://groups.google.com/group/algogeeks?hl=en. > -- Arun Vish Graduate Student Department of Computer Science University of Southern California -- You received this message because you are subscribed to the Google Groups "Algorithm Geeks" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/algogeeks?hl=en.
