OCR for iOS and Android

[Alex40 4]

Hi everyone,

 

I am currently looking for a good OCR library for converting images with text.

 

Can you suggest any other than Winsoft's which is up to date with the latest Rad Studios?

 

I also noticed that the Winsoft's OCR library doesn't seem to support Cyrillic alphabet recognition.

 

Thank you in advance.

[Mohammed Nasman 11]

Hi

ImageEN with it's IEVision plugs for OCR has support to Cyrillic 

 

ImageEn Help 

Also have a look to Object Pascal binding for tesseract-ocr - an optical character recognition engine

[Rollo62 502]

ImageEN is great, but unfortunately VCL only, and Windows specific.

Edited July 12, 2023 by Rollo62

[Guest]

You can use the google vision api, the first 1000 readings per month are free.

you can use this free api, https://ocr.space/OCRAPI
Requests/month 25,000 free.

you can write a backend server,

If your request is a standalone ocr application that will run entirely within the app, I don't think you can get successful results except for recognizing computer fonts, these structures are trained with many images for a long time, so they can recognize handwriting etc properly, and the data sizes are too large, you cannot store and access it on the phone.

I'm using google vision api right now, it gives good results on handwriting and pictures taken at different inclined angles.
I can get the texts and the positions of the texts on the picture within an average of 0.300 ms.

It can be done via a simple ocr python bridge on fmx.

[Fr0sT.Brutal 897]

Tesseract is open source and seems to support Android

https://f-droid.org/ru/packages/org.totschnig.ocr.tesseract/

 

[Alex40 4]

Thank you all for the answers!

Edited July 13, 2023 by Alex40