Post Reply 
 
Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Cameras
Author Message
deljr15 Offline
Administrator
*******

Posts: 102
Joined: Dec 2014
Reputation: 0
Post: #1
Cameras
In the large old thread we had settled on the raspberry pi cameras. We will have 2 available.

1) to be pointed at the moss.
2) has been suggested for use in navigation, and has also been requested by biology for a second viewing Angle.

So what should the second camera be used for?

Ideas
12-20-2014 01:02 PM
Find all posts by this user Quote this message in a reply
Newt Offline
Flying Buttress
***

Posts: 134
Joined: Dec 2014
Post: #2
RE: Cameras
I was wondering about how exactly to know when we are up to the right spin rate. An outward looking camera would enable us to calculate that confidently, by when the Earth/Sun/Moon passes in a circle around. Pretty pictures would not often be available, but before we start the spin we might could take some too. We could also probably do the same thing with an accelerometer, though (pretty pictures aside).
Apart from that there does not seem to be too much (that I can think of) that would be really useful, apart from looking at the moss. It would not hurt to have a second camera there, just in case. There was mention of spectral measurements to determine parameters of the moss' habitat, but I am skeptical as to whether that would be practical with such a camera.
12-20-2014 01:33 PM
Find all posts by this user Quote this message in a reply
MBobrik Offline
Member
***

Posts: 203
Joined: Dec 2014
Post: #3
RE: Cameras
How many cameras we can have is ultimately constrained by the downlink speed, the required number of pictures per day, and by how many ground stations we will have.
If we assume 50 kbps bandwidth and 1 picture per hour per sample, we will need 1 ground station per 3 samples. Doubling the cameras doubles the demands.
It can of course be the case, that the downlink rate is underestimated, or that the timelapse interval can be significantly longer, or both. In this case doubling the cameras would not be a big deal.
Another constraint is, that we need to provide more unobstructed optical paths, which takes volume inside the sat and places further constraints on the container ( we would most probably have to go with the capsule-less design because the capsule walls would obstruct sideways view too much ).
Otherwise than that, there can be as many cameras as we need. One rasp pi compute module supports two cameras, and together with them weights mere 13 grams. We could put two in, for redundancy, which would give us four cameras ( presumably two internal two external ) if we wanted.
.
But @ navigation, I think that while the information from them would be a contribution, it is not necessary, as orientation can be determined from magnetic field sensors and solar panel output, and rotation speed can be directly measured by accelerometers.
(This post was last modified: 12-21-2014 09:04 PM by MBobrik.)
12-21-2014 09:03 PM
Find all posts by this user Quote this message in a reply
Nicholander Offline
Member
***

Posts: 160
Joined: Dec 2014
Post: #4
RE: Cameras
In my opinion, I think 4 cameras is really overkill. I think there should just be 2 cameras, 1 on the outside for figuring out were the CubeSat is pointing and one on the inside for monitoring moss. The inside one should be able to move around the pressurized area by being attached to some motors, so it can see all the moss. Though, then again, you could have 2 cameras with a 180 degree view each so you don't need any motors.
12-22-2014 12:38 AM
Find all posts by this user Quote this message in a reply
Bunsen Offline
Junior Member
**

Posts: 18
Joined: Dec 2014
Post: #5
RE: Cameras
When figuring out how many cameras to include, consider how the cameras can be connected to the computer. The Raspberry Pi (including the Compute Module version, which is probably much better suited to a cubesat) can directly connect to two of RasPi's camera modules. Any more and you'd need to find another connection mechanism -- USB is probably most likely, but it can be surprisingly power hungry.

Also, if this will launch from the United States, there are regulations that require a NOAA license for any space-based imaging system that takes pictures of Earth. Lots of cubesats launched without this license, but they're starting to enforce it now. I don't think it's a showstopper, but it's something to be aware of.

edit: Of course, just as I crap on the idea of USB cameras, I notice this thing: https://www.adafruit.com/products/1722
I don't actually know how useful it would be (obviously we're not looking at photosynthetic activity), nor do I have time to do the background research into what wavelengths it observes or what you can learn from those, so I'll leave that to the botanists to evaluate. Hardware-wise, it looks pretty danged easy to wedge into a cubesat. And the project behind it sells just the filter material, so other RGB cameras can be converted...
(This post was last modified: 12-22-2014 03:20 AM by Bunsen.)
12-22-2014 03:03 AM
Find all posts by this user Quote this message in a reply
Newt Offline
Flying Buttress
***

Posts: 134
Joined: Dec 2014
Post: #6
RE: Cameras
I would be somewhat concerned about field of view for navigation cameras. If it can only view a few degrees, it would be a challenge to point at the sun (or anything else) based on its data. The Raspberry Pi cameras, do not seem suited to that unless we modify them with another lens, which might be possible.

As for the camera Bunsen mentioned in the edit, I have a friend who works on space born evaluation of plant activity, and that sort of thing. I might can talk to him and see what he thinks of that's practicality.
12-22-2014 06:48 AM
Find all posts by this user Quote this message in a reply
MBobrik Offline
Member
***

Posts: 203
Joined: Dec 2014
Post: #7
RE: Cameras
(12-22-2014 03:03 AM)Bunsen Wrote:  When figuring out how many cameras to include, consider how the cameras can be connected to the computer. The Raspberry Pi (including the Compute Module version, which is probably much better suited to a cubesat) can directly connect to two of RasPi's camera modules. Any more and you'd need to find another connection mechanism

Or another rasp pi compute module, as I suggested.

(12-22-2014 03:03 AM)Bunsen Wrote:  edit: Of course, just as I crap on the idea of USB cameras, I notice this thing: https://www.adafruit.com/products/1722
... And the project behind it sells just the filter material, so other RGB cameras can be converted...

Any USB camera would require additional CPU. And the Rasp Pi cam comes in NoIR version which provides equivalent IR viewing capabilities.
12-22-2014 08:13 AM
Find all posts by this user Quote this message in a reply
deljr15 Offline
Administrator
*******

Posts: 102
Joined: Dec 2014
Reputation: 0
Post: #8
RE: Cameras
(12-22-2014 12:38 AM)Nicholander Wrote:  In my opinion, I think 4 cameras is really overkill. I think there should just be 2 cameras, 1 on the outside for figuring out were the CubeSat is pointing and one on the inside for monitoring moss. The inside one should be able to move around the pressurized area by being attached to some motors, so it can see all the moss. Though, then again, you could have 2 cameras with a 180 degree view each so you don't need any motors.

I agree 4 cameras are overkill. Unless we can find very good uses for them, and have a solution that the CPU team can work with.

However I will need to argue against moving cameras. Each camera will need to have a clear viewing area in front of it. Meaning that a moving camera will need clearance for itself to move as well as clearance in front of it at every position.
12-22-2014 11:33 AM
Find all posts by this user Quote this message in a reply
smartS=true Offline
Junior Member
**

Posts: 11
Joined: Dec 2014
Post: #9
RE: Cameras
You might only need 1 camera to record the vertical and horizontal growth of the moss, if it were placed diagonally of the surface the moss was growing on.
12-22-2014 06:33 PM
Find all posts by this user Quote this message in a reply
MBobrik Offline
Member
***

Posts: 203
Joined: Dec 2014
Post: #10
RE: Cameras
My new rasp pi camera's first light.

[Image: EE1GLmt.jpg]

Now I am gonna to experiment with all the stuff we need to set the optics right. Field of view, depth of view, distance, resolution, light sensitivity, and so on...
12-23-2014 01:58 AM
Find all posts by this user Quote this message in a reply
Post Reply 


Forum Jump:


User(s) browsing this thread: 1 Guest(s)