Not Google, but similar.
Yandex Translate has image upload using desktop web browsers.
I use this in a Flotato browser which enables me to quickly translate screen grabs without them accumulating on my hard drive (because the screen grab is automatically deleted after I've dragged it onto the Yandex/Flotato window).
The purely Google approach I use is:
Upload the image to your Google Drive
Right click image and choose Open with > Google Docs
A new document will be created containing the image and extracted text
In Google Docs choose Tools > Translate document... and choose your target language
A second document will be created with the translated text
It has moved elsewhere. OCR is no longer performed during upload.
Instead, upload your files using "drag and drop", or the red "New" button (top-left of Google Drive Web UI). After uploading, right-click the file and choose "Open with" --> "Google Docs". Drive will convert the image to a Google Doc, and when you open the new converted file, the image ...
First, change your Google Drive setting, if necessary (maybe back again afterwards), to check Convert uploaded files to Google Docs editor format (see).
Upload your image file to Google Drive, open it and choose Tools > Translate document... and the language of your choice (from those available!) then click Translate.
As an example I selected a snip (.JPG) ...
Yes, it does. With my digital camera, I took a picture of one page of a book I'm reading. I then uploaded the JPG file to my GDrive. I immediately searched for a word found in the book. It found nothing. After 2 minutes, I tried again, and this time, GDrive found the word. (I guess it takes a few minutes for GDrive to finish OCRing images.)
Now that you uploaded a scanned PDF (or an image file), do this:
Browse to the GDrive web interface.
Right-click on your uploaded image file.
Invoke the Open with > Google Docs menu command.
See the relevant Google Drive help page for details and tips. E.g. the text must be right side up and in a common, readable font; the file should be 2 MB or less, ...
Download the book as PDF (from the "gear" menu on the book's page in Google Books), and open the file in a PDF reader such as Adobe Acrobat Reader or Apple Preview. From there you can Select All, Copy, and Paste into the word processor of your choice.
The Translatium desktop app allows image upload.
You may be lucky to find a download of this app, because it was discontinued. Why? The unofficial API it uses can sometimes fail. My solution is to simply try again a minute or two later and it usually goes through OK.
If you are not restricted by the filetype, and can generate image files of the Scanned documents, then you should take a look at Project Naptha
It is a Chrome extension that performs an OCR on images, and lets you copy (and even edit) the text. I've tried it on images on my Drive Image Viewer, and it's a breeze. You directly have the pointer change to a ...