Ask Your Question

Revision history [back]

Detecting object in video stream in real time and using OCR for text extraction

I'm looking for a way to detect a business card in a video stream (just like human face detection) and extract the email and phone number using once the business card is detected.

I'll use Tesseract for text extraction, my main problem actually is to be able to recognize the card, i've did it through Sift and Surf: they onverconsume the cpu and they have poor training to detect multiple cards.

I'm looking for a way to make the detection very quick and fluid.

Thanks !