Search code examples
google-cloud-vision

OCR using google-cloud-vision - Result does not contain uni characters for Polish, German, etc

I am trying to use OCR feature in Google Vision API but not able to receive expected result. I expect to see ü for German and ć, ń, ó, ś, ź, ł, ę, ą for Polish in the results. Is there a way I can do it?

Obtained text does not contain uni characters for many languages: Polish, German. But this languages in the list of supported languages and language was detected correctly.

enter image description here

I use drag&drop option here https://cloud.google.com/vision/ and CloudVision Android Sample. Thank you for any advices.

Solution

  • I solved this problem. For gitting UNI characters in the result you need to set LanguageHints.

    For Java it will be:

    ImageContext imageContext = new ImageContext();
List<String> languages = new ArrayList<>();
languages.add("pl");
imageContext.setLanguageHints(languages);
annotateImageRequest.setImageContext(imageContext);

    Now I have ć, ń, ó, ś, ź, ł, ę, ą for Polish in the results.