Is there a way to recognize text in a graphic or a picture with tesseract?

First, sorry for my English. I have a question regarding tesseract. Is there a way to recognize text in a graphic or a picture without having to clean the image you want to rec开发者_如何学Pythonognize? in what I read there to clean the image first, remove graphics and photographs and leave only the text. But I want the user to upload to the server and newspaper clippings you can recognize this news without human intervention. It can be tricky. But if you know any other way I could do as they are grateful. Thank you very much

---

No, you can't.

Tesseract is made for reading text and only text. When you perform OCR on a subject with both text and an image, Tesseract spits out things it finds in the image (garbled crap).

You can detect image regions and crop them out, though. I think that would be a better question to ask.