I am using google vision API for text detection and I get the expected result for English but when test on Arabic I get no result, I found in google decumentation that it support many languages includes Arabic by using parameter languagehint ar
but i don't know where to use this parameter.
I use this tutorial.
Till the date Google Mobile Vision API does not support Arabic language. Please refer below link
https://developers.google.com/vision/text-overview
Related
I have implemented Google vision ocr from this documentation but I have to translate the detected live text to Arabic.
https://developers.google.com/ml-kit/vision/text-recognition/v2/android
This is the Quick start provided by Google.
https://github.com/googlesamples/mlkit/tree/master/android/vision-quickstart
I was able to install and detect the text in English but not able to translate in Arabic.
I am creating an app which takes the picture and recognize the text from it and provides it as the output. I am using mlkit from the firebase for this purpose. So far I am able to detect the text in English. Now I want it to detect the text in Hindi. Also, I am using on-device detection for detecting the text.
Can anyone guide me with this hurdle?
Any help will be appreciated.
The Firebase overview of the text recognition API starts with this:
With ML Kit's text recognition APIs, you can recognize text in any Latin-based language (and more, with Cloud-based text recognition).
Since Hindi uses non Latin-based characters, it can't be recognized by ML Kit's on-device API. The Cloud-based API that is linked does support many more character sets, including Hindi.
So you'll either have to use the Cloud-based API, or train your own model to recognize Hindi text.
How to extract arabic text from image in ocr libary android app
Try using google vision API: https://cloud.google.com/vision/
This API supports many languages including Arabic.
You can try the API online and upload an image that contains Arabic writing and see the results for yourself.
I'm developing android app using mobile vision api
the docs in vision api says it provide recognize face, barcode and text.
but I hope to find is smog, light and fire that is recognized in web vision api.
is there way to recognize those things in mobile vision api?
I think the best way to get the results is getting result as json.
please help me
thanks!
Google Mobile Vision is limited to recognizing Face, Barcodes and Text.
For the use case that you are looking for, try looking into Cloud Vision API by Google.
It recognizes different kinds of objects and environments and output of it can be obtained in a JSON format.
I'd like to write android application for text recognition in images from camera. However, text to be recognized does not consist of regular words, it's just a sequence of letters, digits, slashes, etc. Is it still possible to use Google Vision API for this task? Or should I look for some other tools? I have read about tessaract library, but Google Vision seems to be easier for beginner developer. Would it give significant boost in accuracy of recognition?
I can confirm that at least the sample OCR reader that google supplies can read non-language text strings. I have been using it to read VIN numbers, for instance.