-2

Firstly, my apologies for being sligtly off topic since this does not pertain directly to progamming, but only programmers would know.

There's a lot of bitmap data floating around the market, which pay you to enter such data into a text format. Apparently the software that generated these can export bitmaps, but not the data directly, requiring humans to manually type it in again.

I did some research online, and it appears both SAP and Lotus 123 can export data, with little or no restrictions. Could it be these or something else? Can you identify the legacy software that created such data?

Brian Tompsett - 汤莱恩
  • 5,753
  • 72
  • 57
  • 129
  • Short answer: No. There are thousands of systems from which this could have come. In fact, the specified image appears to simply be a scanned document. – Dark Falcon Feb 20 '13 at 17:23
  • No its not a scanned document because the text is perfectly inline (not rotated or skewed) – John Hammond Feb 20 '13 at 17:31
  • 2
    Who says you couldn't can something perfectly straight? Or run a deskewing algorithm on the image? If it isn;t scanned, why is there noise? – Dark Falcon Feb 20 '13 at 17:40
  • "Apparently the software that generated these can export bitmaps, but not the data". That's not so apparent to me. It's harder to convert data to an image than to simply dump into a text file. I'd say it's more likely there were business processes involving paper which the company would like to digitize. OCR is not perfect so they are using humans for the task. – mike Feb 20 '13 at 20:06

1 Answers1

0

This looks like a scanned image of a printed paper.

Circumstantial Evidence:
--I blew up the image and the characters like "C" are not identical. This suggests printed ink.
--There is noise around the letters. This suggests a paper was scanned and/or bits of ink.

The evidence does not conclusively prove that it is a scanned paper. But I don't see the point in intentionally generating noise and inconsistent fonts in an image.

mike
  • 329
  • 2
  • 11