While running an online seminar in professional development the other night, I had a request for the text of the Nebraska folk song with which historian Louis Warren concluded his presentation, “Settling with Debt: Western Development in the Railroad Era.” All Louis had with him was a hard copy of the song’s lyrics, which were printed at the bottom of his last page of notes. He was happy to share the text, but he understandably did not want to let go of his notes. So, I snapped a photo with with my phone, focusing on the bottom portion of the page.
This morning, I opened the original image in Picasa for some simple tweaks. First, I cropped out all irrelevant, surrounding text, and then brightened the image and heightened the contrast. The result is a more white background and darker, clearer text.
Next, I uploaded the image to Google Docs. I had read that Google Docs now supports OCR (optical character recognition), and this was my first opportunity to test it. When you upload an image and want Google to attempt OCR, be sure to check the box to convert text in images and PDFs to documents (see below).
The result, as you can see in the image below, is an image in the top portion of the page and editable text in the bottom portion.
Toward the bottom of my photograph, the image bends a little. I’m not sure if this is an effect of the wide-angle lens on my phone or perhaps I did not lay the sheet of paper down flat on a table. Nonetheless, the angled lines of the image cause the OCR process not to accurately recognize the points at which one line ends and another begins.
I went back to the image in Picasa, straightened it, then uploaded it once again to Google Docs. The straightened image produced better results.
To finish it up, all I needed to do was clean up some odd spacing in the text (see image below).
While this folk song presents a simple set of text, an amount that surely would not have been a burden to retype, this sample demonstrated to me the value of an accurate OCR process. I’m happy to have this tool in my belt when I need to take on a larger, longer transcription project.
Hurrah for Lane County, the land of the free,
The home of the grasshopper, bedbug and flea,
I’ll holler its praises, and sing of its fame,
While starving to death on a government claim.
My clothes are all ragged, my language is rough,
My bread is case-hardened, both solid and tough,
The dough is scattered all over the room,
And the floor would get scared at the sight of a broom
How happy I am on my government claim,
I’ve nothing to lose, I’ve nothing to gain
I’ve nothing to eat and I’ve nothing to wear,
And nothing from nothing is honest and fair.
– traditional folk song, Nebraska
This post originally appeared at nicomachus.net.