OCR PDF converter to numbers?

[goofyY]

I need to convert a scanned PDF to a numbers spreadsheet. I can't seem to find anything out there that will do it. I'd be willing to pay a bit for it, but so far, even on some downloaded trail versions, I get a blank numbers sheet.

 

[cruisin]

The main problem is that your scanned pdf is essentially an image (in a pdf). You will need something that does OCR to get text out of the image. This might be why your pdf to spreadsheet converters get a blank result. See http://computers.tutsplus.com/tutorials/5-ways-to-ocr-documents-on-your-mac--mac-49683.

The other issue is that you will get a bunch of text/numbers from the OCR process, not a spreadsheet. Look into csv files, as you can take text, put commas to separate the cells, and Numbers will see that as a spreadsheet (or at least Excel does).

If you are handy with find/replace, making the csv file might be fairly simple. Otherwise, I don't know of any other way to get a spreadsheet from an image.

 

[goofyY]

Thanks. I had seen that article, and was trying ocr but I still ended up with nothing. (not even random numbers). I did purchase PDFpedPro which will scan the pdf document using ocr. It looks the same as it did in PDF but is searchable. I just don't know what to do with it now to get it into numbers.

Maybe I need to get it into a cvs file?

 

[cruisin]

csv (comma separated values) is what you need. Something like this:

Item	Units	Price
A	1	$5
B	2	$10

looks like this as a csv:

Item,Units,Price
A,1,$5
B,2,$10

as a csv.

Select all text (⌘ + A), copy it (⌘ + C), open textedit, and paste the text inside (⌘ + V). Now remove all formatting (Format > make plain text) and save it as blah.csv (the blah can be anything). It will ask you if you want to use txt or csv so select csv.

Now it is just a question of making sure the values are separated by commas with no spaces, and it should work. Hope that helps.

 

[Alexander B.]

Try ABBYY FineReader, it has some capabilities for recognition of tables.

 

[goofyY]

thanks,
it seems the scan was done from greyscale and is too poor in quality to be accurately read