Detect Pages
Images are sometimes not shown due to bandwidth/network limitations. Refreshing the page usually helps.
You are currently reading a thread in /g/ - Technology
You are currently reading a thread in /g/ - Technology
I'm looking for advice with RFID or an alternative.
I'm working on a project where I need to be able to detect the pages of a very large "book" (around 20 pages and something like 1 m2 squared).
I need to be able to detect each page to know where the book is open (ie which page).
This was done with RFID tags and sensors but these can't be too close, so it limits the number of pages (only 3 readers can be placed on each side with the current size, see pic related)
I know RFID readers will fuck up if placed more closely, especially with passive tags.
I could put more and turn each one on and off, but it's advised to do so every 100ms, and I'm afraid doing this too much would make it "slow", as I want this system to be reactive. 100ms doesn't seem like much but if you start doing it for multiple sensors it'll begin to add up.
Does anyone have a suggestion concerning RFID or an alternative method of detection?
It needs to run on a computer and pass through USB, current system is using phidgets sensors so sticking with phidgets stuff would be the easiest
>>55629498
perhaps photoresistive sensors?
or if the system can be separate from the book, try cameras and machine learning/artificial intelligence?
also try >>>/diy/
>>55629636
cameras are not possible as people are going to interact with the book, and the pages will be blank, images will be projected on it.
I need to know the physical page I'm on so I can project stuff on it, the goal is to have an interactive book (with videos/ animations on the pages)
I'll also try on diy, thanks
embed capacitors (or resistors) in the pages such that they click together
they will stack up in a parallel circuit and the capacities will add up
or print a QR code on each page, those are pretty robust
>>55629969
this book is going to be displayed in public and used by hundreds of people. Since the pages are just paper, it can "move around" a bit and such, I'm afraid this method would be a bit risky. Having your hand on it would change everything, especially if you wear a wool sweater or something as you'd be electrically charged.
Printing a QR code would require the sensor to be able to see the page, so It would have to be on top of the page? unless you could read it through other pages (I don't think that's possible,I don't think anyone would let me install an X-ray QR code reader) it won't really work since users have to flip the pages, touch it and interact with it in general
posted it on DIY btw,
>>>/diy/1024486
>>55630153
how would it not work? put a code in the top corner encoding the page number, mount a cheap webcam on the ceiling, zoomin in on that corner (preferrably optical zoom)
run a script that scans the image 10 times per second, it will detect the code even if its shifted or slightly rotated
since people usually tent to read books in upright orientation, the top half is unlikely to be covered, especially if there is a wall behind it
or you could use small magnets instead of RFID, 1m^2 is plenty of room that they wont interfere with each other. get 50 for a few bucks from anywhere on /g/chinkschit, for detectors you can either rig some simple mechanical switch in the base or get a bunch of hall sensors for an arduino or Rpi, they cost only a few dollars
>>55631424
>since people usually tent to read books in upright orientation, the top half is unlikely to be covered
yeah but it's gonna be used by hundreds of people every day, "unlikely" then becomes a certainty
>you could use small magnets
someone suggested that on /diy/ but as I said, I don't want someone to come with a bracelet that's held with magnets to fuck up it up, it might sound picky but lots of people are going to use it and that means there's bound to be a couple of them with stuff like that
>>55629498
Cut holes in a grid and use some sort of sensors to determine the page?
>>55629498
Couldn't you space the sensors diagonally, then fill the opposing corners? It would allow for 4 sensors per side.
>>55629498
Also, could you use a weight sensor? To make it more accurate you could use some balast on every page.
>>55631904
I thought about adding sensors to the top of the page, I'll post a quick MS paint of my idea in a couple of minutes
>>55631917
people are gonna have their arms or hands on it, weight is not gonna work
Two weight sensors, one for each side of the book. You measure the weight distribution between the sides against the total weight.
For example if 75% of the weight is on the right side -> 75% of the pages are on the right side -> you're 25% through the book
If you're gonna do it for just one book then you can hardcode the page count.
"""Schematic"""
w = weight sensor
m = wood or some other material to make the framemmmmmmmmmm
m m
m m
m m
m m
m m
w w
m m
m m
m m
m m
m m
mmmmmmmmmm
>>55631974
it has proximity sensors underneath the pages that act as buttons, people are going to touche the pages, it'll fuck the weight up.
>>55631904
here is my Pro mspaint drawing, I'd check the blue sensors with the red ones turned off and then the other way around, so they don't interfere with each other.
that's what I intend to do if I can't find a better alternative
bumping for interest
