hocr.h man page

src/hocr.h — libhocr main C language header.

Synopsis

#include <ho_pixbuf.h>
#include <ho_array.h>
#include <ho_array_hist.h>
#include <ho_array_stat.h>
#include <ho_bitmap.h>
#include <ho_bitmap_hist.h>
#include <ho_objmap.h>
#include <ho_layout.h>
#include <ho_dimentions.h>
#include <ho_segment.h>
#include <ho_font.h>
#include <ho_recognize.h>
#include <ho_recognize_nikud.h>
#include <ho_linguistics.h>
#include <ho_string.h>

Defines

#define HOCR_H 1

Functions

ho_bitmap * hocr_image_processing (const ho_pixbuf *pix_in, const unsigned char scale, const unsigned char no_auto_scale, double rotate, const unsigned char no_auto_rotate, const unsigned char adaptive, const unsigned char threshold, const unsigned char a_threshold, int *progress)

ho_layout * hocr_layout_analysis (const ho_bitmap *m_in, const int font_spacing_code, const int paragraph_setup, const int slicing_threshold, const int slicing_width, const int line_leeway, const unsigned char dir_ltr, int *progress)

int hocr_font_recognition (const ho_layout *l_page, ho_string *s_text_out, const unsigned char html, int font_code, const unsigned char nikud, unsigned char do_linguistics, int *progress)

int hocr_do_ocr_fine (const ho_pixbuf *pix_in, ho_string *s_text_out, const unsigned char scale, const unsigned char no_auto_scale, double rotate, const unsigned char no_auto_rotate, const unsigned char adaptive, const unsigned char threshold, const unsigned char a_threshold, const int font_spacing_code, const int paragraph_setup, const int slicing_threshold, const int slicing_width, const int line_leeway, const unsigned char dir_ltr, const unsigned char html, int font_code, const unsigned char nikud, const unsigned char do_linguistics, int *progress)

int hocr_do_ocr (const ho_pixbuf *pix_in, ho_string *s_text_out, const unsigned char html, int font_code, const unsigned char do_linguistics, int *progress)

const char * hocr_get_build_string ()

const char * hocr_get_version_string ()

Detailed Description

libhocr main C language header.

libhocr - LIBrary for Hebrew Optical Character Recognition

Define Documentation

#define HOCR_H 1

Function Documentation

ho_bitmap* hocr_image_processing (const ho_pixbuf * pix_in, const unsigned char scale, const unsigned char no_auto_scale, double rotate, const unsigned char no_auto_rotate, const unsigned char adaptive, const unsigned char threshold, const unsigned char a_threshold, int * progress)

convert a gray pixbuf to bitmap

Parameters:

pix_in the input ho_pixbuf
scale the scale to use
no_auto_scale do not try to autoscale
rotate the rotation angle to use
no_auto_rotate do not try to autorotate
adaptive what type of thresholding to use. 0-normal,1-no,2-fine.
threshold the threshold to use 0..100 (0-auto)
a_threshold the threshold to use for adaptive thresholding 0..100 (0-auto)
progress a progress indicator 0..100

Returns:

newly allocated gray ho_bitmap

Referenced by hocr::Hocr::do_image_processing().

ho_layout* hocr_layout_analysis (const ho_bitmap * m_in, const int font_spacing_code, const int paragraph_setup, const int slicing_threshold, const int slicing_width, const int line_leeway, const unsigned char dir_ltr, int * progress)

new ho_layout

Parameters:

m_in a pointer to a text bitmap
font_spacing_code -3 tight .. 0 .. 3 spaced
paragraph_setup free text blocks or boxed in columns
slicing_threshold percent of line fill to cut fonts
slicing_width what is a wide font
dir true-ltr false-rtl
progress a progress indicator 0..100

Returns:

a newly allocated and filled layout

Referenced by hocr::Hocr::do_layout_analysis().

int hocr_font_recognition (const ho_layout * l_page, ho_string * s_text_out, const unsigned char html, int font_code, const unsigned char nikud, unsigned char do_linguistics, int * progress)

fill a text buffer with fonts recognized from a page layout

Parameters:

l_page the page layout to recognize
s_text_out the text buffer to fill
html output format is html
font_code code for the font to use
nikud recognize nikud in image
progress a progress indicator 0..100

Returns:

FALSE

Referenced by hocr::Hocr::do_font_recognition().

int hocr_do_ocr_fine (const ho_pixbuf * pix_in, ho_string * s_text_out, const unsigned char scale, const unsigned char no_auto_scale, double rotate, const unsigned char no_auto_rotate, const unsigned char adaptive, const unsigned char threshold, const unsigned char a_threshold, const int font_spacing_code, const int paragraph_setup, const int slicing_threshold, const int slicing_width, const int line_leeway, const unsigned char dir_ltr, const unsigned char html, int font_code, const unsigned char nikud, const unsigned char do_linguistics, int * progress)

do ocr on a pixbuf

Parameters:

pix_in the input ho_pixbuf
s_text_out the text buffer to fill
scale the scale to use
no_auto_scale do not try to autoscale
rotate the rotation angle to use
no_auto_rotate do not try to autorotate
adaptive what type of thresholding to use. 0-normal,1-no,2-fine.
threshold the threshold to use 0..100 (0-auto)
a_threshold the threshold to use for adaptive thresholding 0..100 (0-auto)
m_in a pointer to a text bitmap
font_spacing_code -3 tight .. 0 .. 3 spaced
paragraph_setup free text blocks or boxed in columns
slicing_threshold percent of line fill to cut fonts
slicing_width what is a wide font
dir true-ltr false-rtl
html output format is html
font_code code for the font to use
nikud recognize nikud in image
progress a progress indicator 0..100

Returns:

FALSE

int hocr_do_ocr (const ho_pixbuf * pix_in, ho_string * s_text_out, const unsigned char html, int font_code, const unsigned char do_linguistics, int * progress)

do ocr on a pixbuf, using default values

Parameters:

pix_in the input ho_pixbuf
s_text_out the text buffer to fill
html output format is html
font_code code for the font to use
progress a progress indicator 0..100

Returns:

FALSE

const char* hocr_get_build_string ()

return the build string

Returns:

build string

Referenced by hocr::Hocr::get_build_string().

const char* hocr_get_version_string ()

return the version string

Returns:

version string

Author

Generated automatically by Doxygen for libhocr from the source code.

Info

31 Jul 2008 Version 0.10.17 libhocr