The OpenOCR has the ability to preprocess the image using Stroke Width Transform, which is capable of removing the non-text pixels from an image.
the system should combine the stroke width transform and ocr to detect the text in natural scenes.
then use translate the text into video to the users in order to inform the context in the scenes.
the more detail information, please contact me and talk more .
the basic skill is: c++, linux , computer graphics , and the knowledge ocr.
13 freelancers are bidding on average €1676 for this job
Hello Sir, OCR and computer vision expert here. Please lets discuss this project with more detail, so we can agree on the steps needed. Best Regards, ITGold.
Hi, I have experience programming with C and python... I have used some OCR tools before, and your project is related to my experience... You can contact me if you need my help