Image to Text, Capture direct from Display Image

greg_brady · February 13, 2018, 5:37pm

Hi Scott, I just purchased Image to Text and am working through how to use it.

I would like to capture an image to the display and then OCR it without generating an image file.

Is this possible?

media.capturePhoto() lets me save the image to a display object

but

imageToText.convert wants a file name to work from.

Is it possible to pass a display object to this function or create another function to work off of a display image?

(I am working towards OCR on the fly)

Thanks for the help, Greg

Scott_Harrison · February 17, 2018, 1:27am

No, you have to convert the display object to an image first.

greg_brady · February 17, 2018, 4:04pm

Well that sucks, a waste of 10 bucks…

rob · February 17, 2018, 4:15pm

There are various display.save(), .capture() etc. API’s that will save the display to a file. Look those up. I think you would need to create a display.newGroup(), add the image to it and use the appropriate capture tool that saves a display group. Then you should be able to use the OCR plugin to process the file.

Probably about 3 extra lines of code (2 if you don’t mind showing the image in your app)

Rob

greg_brady · February 19, 2018, 12:17am

Hi Rob, the control is slow as molasses… saving the image and then processing it slows it down even more. I wanted to do this project in Corona but I have to switch over to Android Studio and Google Vision API to get it done. (my goal is on the fly OCR)