03-15-2013 03:13 AM
Hi
Currently I am using NI-OCR to recognise the english alphabets. But I have a new requirement where I have to recognise the hindi characters. For this purpose I have followed the mentioned steps in the following link
http://digital.ni.com/public.nsf/allkb/81B45058A063420286256DD7007D64F7?OpenDocument
But still unable to train hindi characters.
Please help me in this.
Regards
Giridhar
03-26-2013 02:20 AM
HI
The "OCR" is not able to recognize specific language, because it acknowledges are characters, so if you train appropriate characters (Hindi), your system will be able to recognize them, the easiest way to make the training is to use the "NI-OCR training interface" located in "National Instruments \ Vision or you could also use VBAI (NI Vision Builder for Automated Inspection (AI))
04-03-2013 03:18 AM - edited 04-03-2013 03:18 AM
Hi,
I am not sure how stable will be Hindi OCR even after training, because NI OCR depends a lot from the spaces between the characters, and in Hindi I am not sure there are such. Especially with the long line above the strings.