do_ocr_m ( Character, Image : : NetId : Classes, Confidences )

Classification of a character.

The operator do_ocr_m assignes classes to the Character (characters). Hereby all grayvalues of the surrounding rectangles of the regions will be used. The grayvalues will be taken from the paramter Image. For each character the two classes with the highest degree of network stimulation will be returned in Classes. The corresponding stimulations will be transmitted in Confidences. The stimulation ranges from -1 (rejection) to 1 (highest degree of stimulation).


Parameters

Character (input_object)
region -> object
Character to be recognized.

Image (input_object)
image -> object
Grayvalues for the characters.

NetId (input_control)
integer -> integer
ID of the OCR network.
Default value: 'integer'

Classes (output_control)
string-array -> string
Classes (names) of the characters.
Number of elements: 2

Confidences (output_control)
real-array -> real
Confidence values of the characters.
Number of elements: 2


Result

If the input parameters are correct, the operator do_ocr_m returns the value TRUE. Otherwise an exception will be raised.


Possible Predecessors

traind_ocr_net, trainf_ocr_net, read_ocr_net, connection, sort_region


Alternatives

do_ocr_b


See also

write_ocr_net



Copyright © 1996-1997 MVTec Software GmbH