It's tough to do. They could try OCR but I think it would be too CPU intensive given the volume of posts Reddit receives. Even if OCR works at first it would be easy to use a harder to read font or apply some effect to the font to reduce OCR accuracy.
OCR means Optical Character Recognition. The goal is to take a picture of text and convert it into text. Depending on the software it can take a little while to process one image. For this reason I'm not surprised that Reddit isn't using OCR on all images posted because there are too many.
It's tough to do. They could try OCR but I think it would be too CPU intensive given the volume of posts Reddit receives. Even if OCR works at first it would be easy to use a harder to read font or apply some effect to the font to reduce OCR accuracy.
Way over my head Pede but I was thinking after I posted that they probably couldn't.
You know. Because of the pixels.
OCR means Optical Character Recognition. The goal is to take a picture of text and convert it into text. Depending on the software it can take a little while to process one image. For this reason I'm not surprised that Reddit isn't using OCR on all images posted because there are too many.