Receiving an error message, unexpected output, or undesirable output when using Grab Text, or Edit Text in Snagit 2018 and later.
Snagit utilizes the Abbyy OCR Engine to recognize and capture text. Below are some known issues, and best practices for using the OCR functionality in Snagit.
How does Snagit determine which font to use?
The Abbyy engine will identify which family font belongs to, and then use a common font in that same family for text replace. For example, if the font detected is determined to be a Monospace font, then the font used when replacing the text will be Courier New. The table below describes the families of fonts detected, examples, and the font that Snagit will use for each scenario.
|Font Family Detected||Example Font||Font Used in Snagit|
|Serif||Century Schoolbook||Times New Roman|
Slanted/angled text may not be recognized
The OCR library that is used in Snagit does not recognize text that is slanted as in the example below. As a workaround, try rotating the image, and then trying the capture again.
More contrast, the better
Light colored text on a light colored background can be difficult to recognize by the OCR engine. For best results, try to work with content where the contrast between the text and the background color is high.
|Light text, dark background||Light text, light background|
|Dark text, light background||Dark text, dark background|
The first OCR function takes a long time
The very first OCR function (Grab text or Edit text) will take longer to process than subsequent captures. After the first OCR function of a Snagit session, it should perform normally.
Fonts and characters that may have problems
Fonts with the following properties may have unexpected results:
- Glyph or symbol based fonts
- Fonts with accents
- Handwritten or cursive font
Make sure the font is large enough
If a font is too small, then Snagit will not be able to detect that text exists. As a general rule of thumb, if the font is readable on the screen, Snagit should be able to capture it.
Bounding boxes surrounding fields of text
Text that is in a bounding box may prevent the OCR engine from recognizing the text itself. As a workaround, use the Selection tool inside of Snagit to select only the text within the bounding box, and then choose Edit Text or Grab Text.
Outlined and/or shadowed text
Shadowed text or shadowed text may have problems capturing. The text may either be inaccurate, or not recognized at all.
Grab text on large captures
When using Grab Text on a very large capture (for example, a scrolling capture or panoramic capture), it may take Snagit a very long time to process the capture. We do not recommend using Grab Text or Edit Text on very large captures. Instead, use the Selection tool to select a smaller portion of the capture to work with.