By Unregistered User (not verified), 7 December, 2012
Forum
iOS and iPadOS
Dear AppleVis,
I am interested in using Prizmo with students with visual impairment and
blindness. Is it possible to bring in a .pdf file from Dropbox into Prizmo to
turn into OCR format? Please advise.
Thank you.
So far as I can tell, no it is not possible to do this. I am a student and would like to see this feature. Can anyone suggest an app that will do this?
Hi,
It is not currently possible to OCR from PDF files with any of the iOS apps that I know of. The best you can do, on a device with retina graphics, is to take a screen capture of the displayed PDF images, and then use an OCR app to process the screen image from the camera roll. Prizmo can do this, but there are a number of limitations.
1.You would have to have the screen curtain off, and the brightness of the display set to at least a minimal value which would be around 30% or higher. While using the camera doesn't require the screen curtain to be on, taking a screen capture image, by simultaneously clicking (i.e., short press and release) on the Home button at the bottom of the screen and the sleep/wake button at the top right of the device, does require the screen display to be visible, or your starting image will be solid black and produce no results. You should hear the sound of a camera shutter when you release the buttons for a screen capture. Just make sure to press the home button slightly before you press and release the sleep/wake button so you don't end up locking your screen instead of doing a screen capture. For multiple page PDFs you would have to screen capture each page separately.
2. You are taking a picture of a picture, and the quality of the image will be degraded by the additional step, and also depends on the device/camera you're using. So how good the results are may depend on whether you are using an iPhone 5 or Retina iPad vs. other devices.
3. If you do work with screen captures in Prizmo, after you take the screen capture the instructions are simple:
3a. On the main screen, flick past the settings button to "Text" and double tap
3b. On the photo screen, flick right past the "get text from picture" announcement and past the "camera" button to the "album" button and double tap. (Before you do the double tap, note that the bottom of the screen has 3 tabs, and that currently the first of these, the "source" tab, is selected.
3c. On the photos screen, flick right to "camera roll" and double tap. (The number of pictures in your camera roll will also be announced.)
3d. On the camera roll screen, do a four-finger tap on the bottom half of the screen to move focus to the last picture, then double tap to select it.
3e. On the Photo screen, flick right to the "Next" button in the top right corner and double tap. (Again, before you double tap, note that the second "Image" tab at the bottom is selected.)
3f. You'll hear "processing" and when you flick right you'll be on an "Edit" screen, where you can flick right to read, and possibly correct typos in the results.. Or you could just double tap the "Next" button in the top right corner again and be taken to the "Text" screen, where you can also read, but not edit the text.
3g. On the bottom of the "Text" screen are button options. Ignore "save", "read", and "translate" . "Save" keeps another copy of the image in the app -- this is mostly for trying different processing options where you visually adjust or crop the image, and will eventually take up lots of space. "Read" prompts you to buy one of the Acapela voices for reading out the text. "translate" uses Google Translate to translate a document scanned in another language, The likely options to use are the "copy" button, which copies the text to your clipboard for pasting elsewhere, or, if you do a three finger flick right to scroll to the second page of buttons, "mail", "open in", or "Dropbox". Make your selection then double tap "Done" in the top right corner to exit,
4. If you really want to try an OCR app on screen captures, I'd recommend instead getting TextGrabber by ABBYY (the same people who make the ABBYY FineReader Express OCR software for the Mac.) It's many fewer steps, even if in Prizmo you're just double tapping the "next" button in the top right of the screens. The only configuration step I would make the first time you use TextGrabber is to flick righ to the "Settings VF" button , double tap, then on the "Settings" screen, flick right to "Enable crop" and double tap to set this switch button to "off", since you won't visually adjust the crop. Then double tap the "Done" button in the top right corner to exit setting. Then, as before, take your screen capture.
4a. Launch TextGrabber, and on the first screen, where VoiceOver announces "Viewfinder image", touch the "album"button in the bottom right corner and double tap.
4b. Just as in 3c and 3d above, double tap "camera roll" and move to the last picture. (Here the last element on the screen is the number of images, so flick left to go to the last image after doing your four finger tap on the bottom half of the screen to move to the last element. then double tap.)
4c. The OCR starts immediately (because you turned off the switch for "enable crop" in settings, no user adjustment of image is expected here). You'll hear "Camera" button, which is the way to go back to the first screen. Flick right to hear the date and then the text results, You can double tap to edit, or immediately flick right to the "menu" button in the bottom right and double tap. Flick left to hear the button options: "dictionary", "message", "copy", "email", "search go", "Evernote", "Facebook", "Twitter"' or flick right to get back to the "menu close" button, or buttons for "history" or "translate". Use the home button to leave the app.
If you have a wireless scanner or a wireless or AirPrint printer with a scanner, another option is to use the VueScan Mobile app by Hamrick Software. This lets you start a scan and OCR from the app on an iOS device on the same wireless network. It doesn't give you all the options of an OCR program on a computer, but it will let you start a scan from the app, and get the OCR results on your iOS device. There's a free version of the app that you can use to try out the app, to check whether it works with your scanner, but it won't support the high resolution scanning option that comes with the paid version, and that is needed to get reasonable OCR results. There's also a long list at the developer's web site of supported HP, Epson, and Canon scanners that work with this app. And it can append pages to a scan result.
HTH.
Esther
I use a service called roboBraille to convert PDFS for me. I'm not sure whether it works on IOS, as it requires one to upload files. I should really try that. Anyway, I typically upload my files from my computer, and the service e-mails me the final document. This, at least, works fine on the iPhone. You can try it at
www.robobraille.org
Comments
Importing PDFs
Current iOS OCR apps all use images, not PDFs
I'm not sure about IOS but robobraille