If you visit https://cloud.google.com/vision/ and scroll down a bit you will see the section
Try the API Drag image file here or browse from your computer
If I drag and upload a Kannada (Indian language) .png
file it reads it and shows text.
Can I know the underlying query string or URL so that I can pass the image files through my script?
PS: Google Cloud Vision API doesnt support Kannada language but it is working on this drag drop interface only. So I want to use it.
Any help would be highly appreciated.
I have been performing some tests with the image you provided and the Cloud Vision API, and it looks like the Kannada language is not yet fully supported by all functionalities in this API. However, the DOCUMENT_TEXT_DETECTION
feature type recognizes the text and transcripts it properly.
You can try with your example image using the Google APIs Explorer, which you basically need to populate using the appropriate feature type and the public URL (or Google Cloud Storage bucket) where it is accessible. If you follow the link I shared, you will be able to see directly the results from your image if you execute the request. Below there's a short sample of the kind of response you will be getting.
{
"responses": [
{
"textAnnotations": [
{
"locale": "kn",
"description": "110 CLUUU (ಸಾಮಾನ್ಯ) ವಿಧಾನಸಭಾ ಕ್ಷೇತ್ರದ ಮತದಾರರ ಪಟ್ಟಿ
...
So once you have tested that Cloud Vision API does support Kannada language, you will be able to call the API using a POST
request such as this one:
POST https://vision.googleapis.com/v1/images:annotate?key={YOUR_API_KEY}
{
"requests": [
{
"features": [
{
"type": "DOCUMENT_TEXT_DETECTION"
}
],
"image": {
"source": {
"imageUri": "<PUBLIC_IMAGE_URL>"
}
}
}
]
}
You can also use the suitable Client Library for your preferred programming language, instead of performing the API calls through HTTP requests.