traind_ocr_net ( Character, Image : : NetId, Class, Iteration, Reduction, Rate : Error )

Training of an OCR network by the input of regions.

The operator traind_ocr_net trains the network directly via the input of regions in an image. Any number of regions of an image can be passed. For each character (region) in Character the corresponding name (class) Class must be specified. The grayvalues are passed in Image. The parameter Error provides information on the convergence of the network when the procedure has finished.


Parameters

Character (input_object)
region(-array) -> object
Characters to be trained.

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

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

Class (input_control)
string(-array) -> string
Class (name) of the characters.
Default value: ''a''

Iteration (input_control)
integer -> integer
Maximum number of iterations.
Default value: 10
Suggested values: 1, 2, 3, 4, 5, 6, 7, 10, 15, 20

Reduction (input_control)
real -> real
Factor to minimize the error (terminating condition).
Default value: 0.1
Suggested values: 1.5, 2, 3, 4, 5, 6, 7, 10

Rate (input_control)
real -> real
Learning rate for back propagation.
Default value: 0.01

Error (output_control)
real -> real
Network error.


Result

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


Possible Predecessors

create_ocr_net, read_ocr_net


Possible Successors

traind_ocr_net, info_ocr_net, write_ocr_net, do_ocr_b, do_ocr_m


Alternatives

trainf_ocr_net



Copyright © 1996-1997 MVTec Software GmbH