$extrastylesheet
Functions | |
template<typename I > | |
component_set< typename mln::trait::ch_value< I, def::lbl_type >::ret > | scribo::toolchain::text_in_picture (const Image< I > &input_rgb_orig, bool bg_removal, bool multi_scale_bin, unsigned max_dim_size=0, unsigned lambda=0, bool verbose=false) |
Classical processing toolchains for pictures.
component_set<typename mln::trait::ch_value< I , def::lbl_type >::ret> scribo::toolchain::text_in_picture | ( | const Image< I > & | input_rgb_orig, |
bool | bg_removal, | ||
bool | multi_scale_bin, | ||
unsigned | max_dim_size = 0 , |
||
unsigned | lambda = 0 , |
||
bool | verbose = false |
||
) |
Localize text in pictures.
[in] | input_rgb_orig | A RGB image. |
[in] | bg_removal | If set to True tries to identify background components and remove them. |
[in] | multi_scale_bin | Sauvola's binarization is used. Tells whether to use a multi-scale version or not. |
[in] | max_dim_size | The maximum height or width allowed for the input image. If larger, it is resized. |
[in] | lambda | Maximum area of components considered as foreground objects. If set to '0' (default), it is fixed automatically. |
[in] | verbose | Enable/Disable debug output on std::cout. |