Is it possible "Get Text" in specific language using a different tesseract data?

[romuluc]

Could I "Get Text" in portuguese ou spanish from OCR? Is there any way to calibrate OCR to improve the quality of obtained text?

[edsonharantes]

Yes, you can, take a look at this, you can add a traineddata data with your language and get better results, I'm from Brazil and I did this with portuguese, just look at the docs, don't forget to also configure the test cases to run with this configuration, I think there's a keyword for this.

👍

[rainmanwy]

Thank you @edsonharantes , @romuluc , "Get Text" is using tesseract, so need to add related trainedata.

[YiQiSun-Uniques]

Hi @rainmanwy, I download chi_sim.traineddata and copy it to folder tessdata where eng.traineddata exists in SikuliLibrary.jar, but it still cann't identify Chinese when I use RobotFramework, Do I need to modify other files?

Thanks in advanced

[rainmanwy]

@YiQiSun-Uniques , currently i could not have a try ocr in my environment. Do you check the link as @edsonharantes suggested?

You may find sikulix folder in your home folder.