We closed this forum 18 June 2010. It has served us well since 2005 as the ALPHA forum did before it from 2002 to 2005. New discussions are ongoing at the new URL http://forum.processing.org. You'll need to sign up and get a new user account. We're sorry about that inconvenience, but we think it's better in the long run. The content on this forum will remain online.
IndexProgramming Questions & HelpElectronics,  Serial Library › Totally lost, suggestions
Page Index Toggle Pages: 1
Totally lost, suggestions (Read 1614 times)
Totally lost, suggestions
Feb 23rd, 2008, 5:38pm
 
Hey I have a project task that is rather beyond me at the moment. I'm hoping that someone can point me in the right direction.

I need a way to detect when a piece of paper is being ripped off a day-calendar, and which day. By day-calendar I mean a calendar with a page for each day.

I have looked into wiring and Arduino. I understand these could be helpful. But I am lost as to the sensors and technology and setup I would use to track the data I need.

If anyone has ideas or knows of useful/relevant technology I would be very grateful.

JMMK
Re: Totally lost, suggestions
Reply #1 - Mar 7th, 2008, 9:43pm
 
One thing I'm working on is Background Subtraction. What this does is basically takes a picture of a frame and compares it with any new frame.

If you take a picture (with a webcam...you'll need a live feed) of the calender full and complete, then have people rip off a day on it, you can determine the exact pixels for each day and have the computer/programming/arduino respond accordingly.

There is a sample sketch of background subtraction in Processing.

File/Example/Libraries/Video(Capture)/Background Subtraction.

Hope this helps,
~TSG
Re: Totally lost, suggestions
Reply #2 - Mar 14th, 2008, 12:32am
 
If you're going for a more compact, cheap solution you could use a binary encoding scheme to number the corner of each page, then use several white/black IR sensors to decode the binary pattern (and position it over the corner where the code is).  This could easily be hooked up to a cheap platform like the arduino and send the data to processing via serial.  It'd take nine bits to cover a full year (2^9 = 512).

White/black sensors are only $1.13 at sparkfun, I have a few of them and I know they'd work fine for this purpose.

http://www.sparkfun.com/commerce/product_info.php?products_id=246

They're analog, so you could experiment with reliably detecting shades of gray which would reduce the number of sensors needed.
Re: Totally lost, suggestions
Reply #3 - Mar 14th, 2008, 11:32am
 
Or if you use a camera you could also print a fiducials tag on every page and use reacTIVision.
http://reactable.iua.upf.edu/?software
The only problem there is that your camera has to be quite near the calendar.
Maybe you could use some kind of telephoto lens?
Re: Totally lost, suggestions
Reply #4 - Jun 15th, 2008, 5:43am
 
obviously you use an RFID reader and
http://www.engadget.com/2007/02/28/rfid-staples-omnipotent-pens-to-grace-offices-of-the-future/

=P, i jest...

I can't really think of any sensor that would work ATM - piezos could measure _when_ the paper was ripped off, but definitely not what day was written on the paper... webcam is probably your best bet =]
Re: Totally lost, suggestions
Reply #5 - Aug 8th, 2008, 3:47am
 
I guess that propably your best bet might be not to look for a single, monolithic solution but a combination of several approaches:

Somebody will need to bring their hands near the calendar (=> proximity sensor, maybe capacitative), apply force to it when tearing off a page (maybe best measured where the calendar is fixed to e. g. the wall), and within a certain timeframe after those two events the visible surface will have significantly changed (=> cam).

Just doing background subtraction might fail due to changing lighting conditions (e. g. night/day) and you need to take into account people passing by or standing in front of the calendar. This might be partly addressed by chosing IR lighting and vision, located as near as possible to the object of desire.

In any case you would have to rule out a whole lot of environmental distractions, at least if you are not dealing with a clean laboratory situation. Nonetheless, I would suggest to iterate over the whole thing several times, trying the most simple approach and see for real where troubles arise (which in turn may as well hint at pragmatic solutions) instead of addressing all possible circumstances beforehand. Good luck!
Re: Totally lost, suggestions
Reply #6 - Aug 8th, 2008, 4:34pm
 
Thanks a lot for all the comments. It turned out that other people in my group decided to try another technology to solve the problem. I can't remember the name but you'd all know it. It must be bought and it's often used for visual/audio installations/interactions. We had a version of processing working but alas it kept having memory problems (sooner or later).


Anyways, again, thanks for all the suggestions!

Page Index Toggle Pages: 1