-2

I have followed http://imperialsoup.com/2016/04/29/simple-ocr-android-app-using-tesseract-tutorial/ for making OCR android app using tess-two. It works perfectly for English,Tamil, Bengali and few other languages. However, when I change the language to Hindi, the app crashes. I have put the hin.traineddata file from https://github.com/tesseract-ocr/tessdata/tree/3.04.00

Can anyone guide me on what I'm doing wrong. Thanks in advance.

I am attaching the screenshot from my mobile which is the summary of report bug which I get when the app crashes. Click on the link to view it. Screenshot-1 Screenshot-2

Hars
  • 11
  • 2
  • 5
    What error are you getting? It will be hard to help you without a more specific error description. – cwoebker Feb 12 '18 at 18:00
  • I couldn't figure out the error. So I have attached the screenshots of the error from my mobile. I have edited my question to include the images. Kindly have a look at it. Thanks – Hars Feb 27 '18 at 16:18

2 Answers2

1

Make sure you include all the cube components.

nguyenq
  • 8,212
  • 1
  • 16
  • 16
  • Thanks for your suggestion. I tried putting all .cube files in the same folder where hin.traineddata is located. However, the problem is not resolved. Is there anything else needs to be changed in the code to use these .cube files ? – Hars Feb 13 '18 at 05:24
  • No, that's all I have to do to be able to recognize Hindi text using Tesseract 3.05.01: `hin.traineddata` file and the cube files -- no other settings needed. – nguyenq Feb 13 '18 at 17:29
  • Hi, Thanks for the reply. But I couldn't get it work with .cube files. Have you made it work for Hindi language? – Hars Feb 27 '18 at 16:20
0

I solved it by changing gradle dependency from tess-two to tesseract4android, you can see the latest version from https://github.com/adaptech-cz/Tesseract4Android. And also it works without any cube files.

Bereket Kassahun
  • 51
  • 4