Hey Tom, Thanks for the answer.
I've looked at the links and will do some more research in these directions. However I still have some hope of working with Ocropus because, and I failed to mention this in my original post, the photos of text using my mobile phone are very close in quality to a scanned photo (angles, black-and-white text, lighting). As a matter of fact, if I'm very careful with the zoom and positioning of my phone, I don't even need document analysis to be done, because the output is so clean that tesseract just eats it up. So, I won't bother you with this anymore, I'm just still looking for a good introduction to using ocropus and it's scripting language. I tried "/usr/local/bin/ocropus -h" and found a lot of commands (even got to "segrast") but can't seem to get into scripting mode... The documentation I found is pretty advanced, I'm still missing the first piece of getting it to work after the successful installation :) Once I find out, I'll be happy to write a "for dummies" article if I can have access to the wiki or something... Thanks again, Jon P.S. The "how to install" article in the wiki doesn't instruct to install scons under ubuntu, thought to let you know (I couldn't find how to edit it myself) On Jun 28, 12:00 pm, tmbdev <[email protected]> wrote: > Processing phone images and getting rid of irrelvant portions of an > image reliably is anything but simple. If it were, you'd see people > using handheld cameras for document capture much more widely. The > idea isn't new (I was capturing documents with cameras as far back as > the 1980's); what's hard is the technology needed to make it work > efficiently and under many situations. > > While OCRopus contains a lot of segmentation and image processing > code, it does not try to address camera-capture of documents > directly. We're working on that as part of the Decapod project > (decapod-project.org). Decapod will work with OCRopus. > > You can also see some demos of work we have done on camera-based > document capture here: > > http://www.iupr.com/on-line-demonstrators-1 > > Tom > > On Jun 27, 7:46 pm, Jon <[email protected]> wrote: > > > Hello, > > > I'm rather new to Ocropus. I've managed to compile and install using > > Ubuntu 8.04. > > I'm now starting to look into ocroscript to find out how I can use it. > > > I've done a lot of work with tesseract and I'm able to use tesseract > > to great extent. > > > However, I'm looking to do some document analysis now, and I know > > Ocropus is the way. > > > I'm doing some manipulation on a document photographed with a mobile > > phone, and I'm trying to get rid of everything irrelevant (fingers, > > background, etc.) > > > Unfortunately, documentation of document analysis is scarce in > > Ocropus, so I thought to ask if anyone has a pointer of implementing a > > fairly simple way to crop out everything unnecessary out of a mobile > > phone picture. > > > This is a fairly simple task that I haven't found anything useful in > > the wiki for. I'd love to take my case and write a good tutorial on it > > (the wiki/code project/etc.) > > > Cheers, > > > Jon --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "ocropus" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/ocropus?hl=en -~----------~----~----~----~------~----~------~--~---
