do_ocr_multi

Herror ::do_ocr_multi ( Hobject Character, Hobject Image, const HTuple &OcrHandle, char *Class, double *Confidence ) Herror ::do_ocr_multi ( Hobject Character, Hobject Image, const HTuple &OcrHandle, HTuple *Class, HTuple *Confidence ) HTuple HRegion::DoOcrMulti ( const HImage &Image, const HTuple &OcrHandle, double *Confidence ) const HTuple HRegionArray::DoOcrMulti ( const HImage &Image, const HTuple &OcrHandle, HTuple *Confidence ) const

Classification of characters.

The operator ::do_ocr_multi assigns a class to every Character (character). For grayvalue features the grayvalues from the surrounding rectangles of the regions are used. The grayvalues will be taken from the parameter Image. For each character the corresponding class will be returned in Class and a confidence value will be returned in Confidence. The confidence value indicates the similarity between the input pattern and the assigned character.

Parameters

`Character` (input_object)	region(-array) `->` Hobject: HRegion(Array)
Characters to be recognized.

`Image` (input_object)	image `->` Hobject: HImage
Grayvalues for the characters.

`OcrHandle` (input_control)	ocr `->` HTuple.long
ID of the OCR classifier.

`Class` (output_control)	string(-array) `->` (HTuple.) char *
Class (name) of the characters.
Number of elements: Class == Character

`Confidence` (output_control)	real(-array) `->` (HTuple.) double *
Confidence values of the characters.
Number of elements: Confidence == Character

Example

char Class[128];
long orc_handle;
read_ocr("testnet",&orc_handle);
read_image(&Image,"character.tiff");
bin_threshold(Image,&Dark);
connection(Dark,&Character);
count_obj(Character,&num);
open_window(0,0,-1,-1,0,"","",&WindowHandle);
for (i=0; i<num; i++) {
  select_obj(Character,&SingleCharacter,i);
    do_ocr_multi(SingleCharacter,Image,orc_handle,&Class,_);
  smallest_rectangle1(SingleCharacter,_,&col,&row,);
  set_tposition(row,col);
  write_string(WindowHandle,Class);
}

Result

If the input parameters are set correctly, the operator ::do_ocr_multi returns the value H_MSG_TRUE. Otherwise an exception will be raised.

Possible Predecessors

::traind_ocr_class_box, ::trainf_ocr_class_box, ::read_ocr, ::connection, ::sort_region

Alternatives

::do_ocr_single

See also

::write_ocr

Module

Optical character recognition