Pixel remapping

  Simply put, is to map each pixel of the input image according to certain rules to a corresponding position further up the image, a new image is formed: g (x, y) is the image after the remapping, h (x, y ) function is a function, f is the source image. Can be changed by re-generating a new map image, for example, about the operation of some of the vertical conversion, the image in inversion.

  Horizontal Mirror function function is:

$$\left\{ \begin{array}{l}
{\rm{a}}\left( {x,{\rm{y}}} \right) = - x\\
b\left( {x,{\rm{y}}} \right) = y
\end{array} \right.$$

  Vertical mirroring function features:

$$\left\{ \begin{array}{l}
{\rm{a}}\left( {x,{\rm{y}}} \right) = x\\
b\left( {x,{\rm{y}}} \right) = -y
\end{array} \right.$$

 Correlation function:

. 1 remap (the src, DST, map_x, map_y, INTER_LINEAR, BORDER_CONSTANT, the Scalar ( 0 , 255 , 255 ));
 2  // map_x, map_y, X and Y values as the mapping table, INTER_LINEAR linear interpolation method, BORDER_CONSTANT boundary treatment method.

Guess you like

Origin www.cnblogs.com/fuzhuoxin/p/12144528.html