libeblearn
ebl::rgb_to_ynuvn_module< T, Tstate > Class Template Reference

#include <ebl_preprocessing.h>

Inheritance diagram for ebl::rgb_to_ynuvn_module< T, Tstate >:
ebl::channorm_module< T, Tstate > ebl::channels_module< T, Tstate > ebl::module_1_1< T, Tstate > ebl::module

List of all members.

Public Member Functions

 rgb_to_ynuvn_module (idxdim &norm_kernel, bool mirror=true, t_norm norm_mode=WSTD_NORM, bool globnorm=true)
virtual ~rgb_to_ynuvn_module ()
 destructor
virtual void fprop (Tstate &in, Tstate &out)
 forward propagation from in to out
virtual rgb_to_ynuvn_module< T,
Tstate > * 
copy ()
 Returns a deep copy of this module (abstract).

Protected Attributes

module_1_1< T, Tstate > * norm2
 Normalization for UV.

Detailed Description

template<typename T, class Tstate = bbstate_idx<T>>
class ebl::rgb_to_ynuvn_module< T, Tstate >

convert an RGB input into a YnUVn output, Yn being a Y channel with a local normaliztion, and UV normalized locally together.


Constructor & Destructor Documentation

template<typename T , class Tstate >
ebl::rgb_to_ynuvn_module< T, Tstate >::rgb_to_ynuvn_module ( idxdim &  norm_kernel,
bool  mirror = true,
t_norm  norm_mode = WSTD_NORM,
bool  globnorm = true 
)

Constructor.

Parameters:
normalization_sizeis the size of the kernel used for Yp's local normalization.
mirrorIf true, pad normalization if a mirror of the image instead of with zeros. This can be useful in object detection when objects are close to borders.
norm_modeThe type of normalization (WSTD_NORM by default).

The documentation for this class was generated from the following files: