ocr not working?

asked 2016-06-29 06:01:47 -0500

atv gravatar image

So i installed tesseract, and running opencv3. I'm really excited to try out OCR. I tried out the webcam example in the text module, but all i get is a black stripe (the camera does turn on, see below for output). I also found this great thread about getting at least the ROI, and then maybe use something like gettext (can't use, GPL) to extract the text.


I'm really in the need for something that extracts texts from the webcam, like so : https://www.youtube.com/watch?v=pgth0... https://www.youtube.com/watch?v=Yv6Ik...

Or just by me feeding it a Mat or a ROI. Did anyone have any success with that?


A demo program of End-to-end Scene Text Detection and Recognition using webcam.

Usage: ./ocr [camera_index]

Press 'r' to switch between MSER/CSER regions. Press 'g' to switch between Horizontal and Arbitrary oriented grouping. Press 'o' to switch between OCRTesseract/OCRHMMDecoder recognition. Press 's' to scale down frame size to 320x240. Press 'ESC' to exit.

init done Initializing OCR engines ... [DS] Profile read from file (tesseract_opencl_profile_devices.dat). [DS] Device[1] 1:Intel(R) Core(TM) i7-2620M CPU @ 2.70GHz score is inf [DS] Device[2] 0:(null) score is 1.369032 Done!

edit retag flag offensive close merge delete