SNAGIT OCR

Hello all! We definitely didn't want to remove Text Capture after Snagit 11. I mentioned this in another thread that, but I thought I'd double post here to make sure you all got this info. If there are other aspects of Snagit that you feel have been removed, let us know. We'll try to make it right, show you the alternative in Snagit now, or at the very least tell you what went into making that decision.

Some history here for Text Capture. Windows doesn’t allow us the same level of access as it once did, which made the old Text Capture implementation fail for too many use cases. It wasn't just that that led to a poor experience for most people. It was also how other apps were creating their UIs. Snagit 11 used a technology to get the text directly from the app you captured. There are more apps today set up in a way that makes grabbing text from them with that same process really rough.

We have been looking for a way to bring it back for quite some time. Historically, OCR hasn't been particularly reliable when processing screen shots, it was made for scanning text on paper. There were far fewer pixels in a typical screen capture. Traditionally 96 or 72 dots per inch. Where a typical scanner processes images and brings them in around 300 dpi, and it had more pixels to recognize the characters in a document.

But OCR tech has improved as well as people using higher pixel density monitors. So after evaluating several OCR engines and tuning them for screenshots, we got to a place where we had the results we wanted and found the ABBYY OCR engine to work the best for our customers' use cases.

What's great about this is that we're able to bring this back in a way that is unlikely to break in the future. Considering that we're processing the image itself and no longer relying on other apps or the OS to tell us what text they contain.