I am working in a External HDD manufacturing company. We have product label on each External indicating the serial number, Part Number and Name of the product. We are manufacturing different models so the product name is different for each models. I just want to ensure that correct product label is sticked for the correct model.

So my idea was to capture the text on the product label and make computer to identify the text from the picture and verify it with the data in the server. This way I can make sure about the quality. But I don't know how and where to start. So I need help on this issue and direct me in right direction.

1 Answer 1


You should start from tesseract-ocr (former known as HP Labs OCR). I had used it for captcha for web automation long time ago. The accuracy is very high given that you have a good image quality. The image quality should no be a problem in your case if you provide good lighting on the label.

Your setup would be capture the image to a local file, then use tessearct-ocr to read characters for you in specific regions.

  • Hi @neo Thanks for the suggestion. This software was useful. Now I have some thing to start with. Now I am trying to merge Tesseract-OCR with a VB application to make it user friendly for production.
    – Dragonborn
    Commented Aug 14, 2015 at 3:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.