way to use it for local image about ocrspace

[ErikBoesen]

You could accomplish this quite simply, as follows:

import ocrspace

api = ocrspace.API()
text = api.ocr_url(message.attachments[0].uri)

text will now contain the text from the image.

[FASTANDEXTREME]

I think by-default it uses OCR Engine-1 and I want to use 3 so it would be like this?

api = ocrspace.API()
OCREngine=3
text = api.ocr_url('https://upload.wikimedia.org/wikipedia/commons/8/8f/Arabic-script.png')
print(text)

i want to change engine as it seems current engine cannot support few languages. Also it does not auto detect the language it seems and only extract text from english text images. Is there way to extract text from image of any language that it supports?

[ErikBoesen]

Close:

api = ocrspace.API(OCREngine=3)
text = api.ocr_url('https://upload.wikimedia.org/wikipedia/commons/8/8f/Arabic-script.png')
print(text)

[FASTANDEXTREME]

Close:

api = ocrspace.API(OCREngine=3)
text = api.ocr_url('https://upload.wikimedia.org/wikipedia/commons/8/8f/Arabic-script.png')
print(text)

it still cannot extract text from any other language that it support except english

[ErikBoesen]

Sorry, missed the second part of the question. You can specify your desired language as follows:

api = ocrspace.API(language=ocrspace.Language.Arabic, OCREngine=3)
text = api.ocr_url('https://upload.wikimedia.org/wikipedia/commons/8/8f/Arabic-script.png')
print(text)

See the full list of language options here: https://github.com/ErikBoesen/ocrspace/blob/master/ocrspace/main.py#L4-L28