Junior can read!

I will write up how this MAGIC all worked but check it out... It WORKS!


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
GroG's picture

Ahahah .. yes awesome magic

Ahahah .. yes awesome magic ;)

It be nice to post diagram (ie go to the gui tab and spread out all the little boxes - the screenshot and post the pictures)

There a tesseract service if I remember correctly, and a filter in OpenCV .. are you using the filter ?

kyle.clinton's picture

The short answer is, Yes!

I instantiate the opencv filter but interact with it directly like a service. It seems to work easier to use capture frame which returns the image file name and then tesseract.ocr([image file name]) which return a text string of the identified words and the feed it in to MarySpeech.
I will work on some screenshots and attempt to upload some more code. Thanks!

moz4r's picture

I love it ! Do you think

I love it ! Do you think tesseract can parse other languages ?

kyle.clinton's picture

Good Question Moz!

It is just recognizing characters, it doesn't really care what the characters spell. I am guessing if it has patterns for the characters then it should be able to "read" it. I will look in the documentation on tess4j I am guessing it is not English only.