I'm a historian and have just returned from a trip to the US with about 36,000 photographs (which look more like scans) of documents, mainly letters, from archives across the country. I was wondering if anyone had any suggestions about how to best process the documents in Lightroom the maximise the accuracy of the OCR app (Abbyy Finereader) that I'm using. Basically what I want is to develop a preset that roughly reproduces the effect that an iPhone app I have (Scanner Pro by Readdle), applies to images. Heres an example from that app.
Example 1: https://ibb.co/eVQrCS
Example 2 [processed]: https://ibb.co/fAmysS
As you can see, the app flattens the image and completely eliminates any variations in light + makes the text considerably clearer (this is not a perfect example as the text is already clear in the first image - but it gives you the idea). Is there any way to reproduce this effect in Lightroom? Or can someone suggest a laptop app that can produce a similar result (preferably one that can bulk process images)?
The app also deskews the images, although that's not essential as most of my photos/scans are correctly skewed
I don't think there's any way to do that as well as an app that has algorithms specially designed for text processing. Probably the best you can do is to define one or more presets for your common lighting and exposure situations that could get you reasonably close most of the time.
Below is an example of LR settings that get you pretty close for the particular example you posted: Contrast +100, Highlights +100, Shadows +100, Whites +100, Saturation -100, Transform Auto. You fine tune this with the Tone Curve, but the advantage of the settings in the Basic panel is that they are "image adaptive", adapting to a wide range of exposures and histograms, whereas Tone Curve settings are more specific to particular histograms.