This question already has an answer here:

What I'm trying to do is essentially:

  1. I'm given a picture of a book/paper. All the words in this picture were typed, but I cannot extract the words with my cursor.

  2. From this picture, I want to extract all the words. Is there a program that can take in the picture and extract all the words (with spaces, caps, punctuation, etc...)?

I think my question may be different from the link in the comment below because the picture I am trying to extract words from is a phone picture of a page in a book. The angle is slightly skewed. I have tried Adobe's OCR reader and it does not give me a correct output.

I primarily use macOS. However, I can get access to Windows 7 (and 10) and Linux. Quality is primary, and cost is secondary to an "extent." I'm really look for a good software and I'm willing to spend some money.

I need to handle 100 pages total.

marked as duplicate by Mawg, Nicolas Raoul Dec 21 '17 at 13:27

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

  • 1
    Do you have a lot of it? For "just a few" you may find an online service is easier that doign it all yourself - I've successfully used onlineocr.net to convert screen shots of programming code (java) to plain text that I could then actually use... – ivanivan Dec 21 '17 at 2:28
  • 2
    I have quite a few right now, and I am certain that I will have more in the coming months. Probably somewhere around 100 pages total in the next few months. I tried the adobe OCR tool, and it does not do a good job at all. I tried it on a picture earlier and it gave me a lot of weird looking symbols. – lawson123 Dec 21 '17 at 4:54

One of the key OCR programs is Tesseract OCR which is a command line, open source, Apache licence, cross platform tool. It is free and can support a large number of languages, (>100 including Unicode support) and can be trained for more. It has been developed since 1985 and is still, (2017), being actively developed. It is available from source or as a pre-built binaries for Windows, Linux, Mac, Android.

As with most OCR software the accuracy will depend on the image quality, (which you may be able to clear up), and possibly on training the package for the font, etc. as well as the language. For example, if the angle of the text is skewed it is worth using an image editor such as GIMP to straighten the angle of the image.

There are a number of GUIs listed here under various licences and for various platforms.

Web Services

There are also a large number of web services available that use the same technology but a word of caution - do not use web services for proprietary or confidential information as the use of some such services count as you publishing the information to the world and/or some services reserve the right to use your information in any way that they choose.

  • "count as publishing the information" meaning? Grammatical? – Pacerier Apr 4 at 5:02
  • @Pacerier - I think that I have improved the caution both in terms of clarity and the grammar. – Steve Barnes Apr 4 at 5:13

If you paste an image into MS OneNote and right-click on the image, there is an option for "Copy Text from Picture". You can then paste it where you need it. This method is far from perfect. It depends on the quality of the text in the image and the font used. But, I have extracted long lines of text from images. It can be faster than copying/typing while reading text from an by just typing.

Note: I am using the Windows version.

  • This seems to not work on the Mac version of OneNote. I will try on a Windows version when I get access. – lawson123 Dec 20 '17 at 21:31
  • OK - I added that in the answer in case anyone else looks at this. Best of luck. – Planet.Megan Dec 20 '17 at 23:33

Not the answer you're looking for? Browse other questions tagged or ask your own question.