Problem of my OCR problem when testing a code-text of a BMP picture?

[gpibssx]

Pls see the attachment! It is OCR project. I want to test the regional code of DVD(see the test picture). In the picture, u can see the code in number such as SD542MDV030219-442-171N.i need to test the number is right or not. SO i programm a OCR VI.i use the "NI OCR Training Interface" to generate my Character Set File, named "DVD regional code.abc".But the problem is that when i run my vi,i only can catch one number every time such as "4",if i want to catch several numbers together such as "442", it's fail. I only catch one. The ROI is manual operated.Pls help me ! Thanks a lot!

[Yusuf C]

Hello,

Your character set file is configured for 'Dark on light' characters, which is the opposite of what you have. You need to train all your characters as 'Light on dark' since your background is black. This setting is located on the Threshold tab of the training interface.
Let me know if you have further questions regarding this.

Best regards,

Yusuf C.
Applications Engineering
National Instruments