(Archived) Text Recognition in PDFs for Premium accounts--or Alternativ

[snsok 0]

I upgraded to Premium mainly to have my handwritten notes both organized and searchable. After doing some reading here and in the Blog, i see that only printed text in PDFs becomes searchable. Here are a couple questions:

I notice that even carefully printed handwritten block letters are not searchable. Is that truly the case?

I need a solution that allows me to scan multiple pages at once, as you can do with scanners that do PDFs, but produce a handwriting recognizable file. I guess I have to do a multi-page TIFF for this. Is that correct? Does anyone know an easy way to convert mult-page PDF to multi-page TIFF?

Does the compression of a TIFF file make a difference? Must it be uncompressed or is LZW compression OK? Is there an optimal minimum resolution that can be used so that file size can be minimized?

Also, I take notes on lined paper and graph paper. Does the faint background image of the lines interfere with handwriting recognition? How about paper that isn't plain white?

[engberg 89]

We don't process anything within .TIFF files ... these would just be stored like a .ZIP file or some other unsupported type.

Our JPEG processing libraries do a lot better with handwriting than our PDF system, so you may want to use JPEG if you are relying on handwriting.

[snsok 0]

Well, Evernote just located a handwritten word ("invoice") within a .TIF file. This worked from both my Mac client and the web client.

[engberg 89]

If you drag that image out of your client onto the desktop and inspected it there, I think you'd find that the actual image is a PNG or a JPEG. (Perhaps the client automatically converted a TIF file you added ...)

Thanks

[snsok 0]

Actually, as I drag it to the desktop, it still is a TIF file. You may be right, tho, that some TIF files have a low-res JPG embedded that allows some programs to preview the image. Perhaps that's why this TIF got recognized, tho many of my others aren't.