Project Naptha software is promoted as being able to manipulate text found in images, found in comics, document scans, photographs, posters, charts, diagrams, screenshots and memes on the internet. This is how he Developer describes his product:
“This entire webpage is a live demo. You can watch as moving your cursor over a block of words changes it into the little I-beam. You can drag over a few lines and watch as a semitransparent blue box highlights the text, helping you keep track of where you are and what you’re reading. Hit Ctrl+C to copy the text, where you can paste it into a search bar, a Word document, an email or a chat window. Right-click and you can erase the words from an image, edit the words, or even translate it into a different language.”
Sounds great and I can see some real practical applications for this in my work. Project Naptha is now only available for the Chrome browser. A quick trip to the Chrome Store, and the Project Naptha extension is installed in my browser and now to try it out.
My first attempt to use Project Naptha involved copying some text from a jpg photo. The actual passage I used is above. Note that the algorithm uses blocks of text to copy, so you could have more than one “block” in a sentence. To copy, I could either have used Ctrl+C, or right-click on the box and choose “Copy Text”. The drop down box options are pictured below:
I then pasted the clipboard content into a text writer, FocusWriter. Here is what I got: The pfiorit of anp addilt is lo aaeslhelíze (…. The OCR (Optical Character Recognition) software did not do a very good job, so I went into the “Language” option and changed the OCR engine from “English Ocrad.js” to “English Tesserect”, and I then got this response when the text was copied: The priority of any addict is to anaeslhelize Though this result is better, it still did not get the last word in the block.
Thinking that the font used in the graphic might be part of the OCR issue, I found another graphic, and copied from it, the block of text above. Here with either OCR engine, I got this response: Be somebody who makes everybody feel like a… Success!!! I replicated this test on a half-dozen different graphics and came to this conclusion: The original font used appears to be a major contributor in the level of success for the OCR outcome in this program. I was able to edit text in graphics, delete text, and so on, again, depending on the original font used.
To me, the OCR is too buggy at this point in the development of Project Naptha to make text manipulation in graphics fruitful, with the exception of being able to delete the text altogether. This last feature has many applications, and is the reason Project Naptha is still residing in my browser.