Request a Personalized Demo

Learn how to accurately convert your designs with Scan2CAD

  • This field is for validation purposes and should be left unchanged.
  • Join thousands of happy customers worldwide

Tips for converting poor quality image text

Video Transcript
00:00: Hi, everybody. Luke from Scan2CAD here. So in this video, I’m going to show you an image in which the text isn’t ideal for converting to vector text. This is the image. We can see that we’ve got very jagged characters. It is not what you would consider clean, quite pixelated. We can also see that there are characters touching together, mainly the P and the L. So, I’m just going through the view options here, and I’ve gone to Raster edit mode, and I’m clicking OCR, then settings. And I can see that the character size settings are currently set at a maximum of 12. If I go again and hit OCR and then character size, and I’m just going to select one of the characters here. You see that Scan2CAD now gives a box around that character, and that’s showing that it sets the character to a size of that area. You can see, if we go back into the previous window, a maximum character size of 24.

01:04: So the first thing I’ve done is I’ve hit OCR, and that’s converted, as you can see, a pink version of the text we’re looking at. This pink version is vector, and it’s fully editable. But we see a couple of question marks, and question marks basically represent areas in which Scan2CAD isn’t 100% confident of what the character should be. So we see that an M has a question mark, and this P also has a question mark. And it’s actually because this one’s next to a touching character, the P and the L are touching. So any application will just see that as one character, one element. So to address that,the first thing I will do is undo the OCR, and that basically closes that vector text.

01:52: Next, I want to split the character. So I go to OCR, then settings, and you can see here there’s a touching character’s box, and they’re split. We can hit the check mark next to it, and hit okay. That will tell Scan2CAD to separate those characters. So, let’s hit OCR again. We now see that Scan2CAD has recognized that there should be two P’s there, and it’s converted it as we want to do, even though the original version has touching characters.

02:25: So I want to edit this text string. I am going to the vector edit mode, hit vector draw button, that brings up the vector edit options. I’m hitting the pink A to edit the vector text. This brings up this window in which we can directly edit the character string. So I’m going to change that to an M, hit okay. We have a vector representation of the original Raster image. I hope that helps. Feel free to send any questions to the Scan2CAD support team, we’re here to help. Cheers.