Color Calibration

This is copy pasted from an email and may be a little rough. Can clean it up later when someone has gone through it.

Camera Settings

make exposure low. The trade off here is between high exposure giving motion blur, and low exposure being dark. So basically make exposure as hig as you can without adding motion blur.
make gain maximum. Gain artificial brightens the image, but introduces noise. We dont analyse the image in enough detail to notice the noise and the added brightness helps make up for the low exposure
white balance affects how white looks. Its a bitch to set with the camera driver atm so just make it some constant value and dont stress
saturation affects the amount of colour in an image. There should be a point (~200) where the ball changes to a much darker red, but the wastebands do not. Use this setting to maximise the difference between ball and wasteband
dont remember what contrast does technically, from memory it also helped a little with distinguishing the ball and wastebands
turn off any auto settings

Connect the robot to your computer directly using an Ethernet cable. It is hard to get vision data over the wireless.
Start runswift on the robot and offnao on your computer.
Connect to the robot by going File -> Connect to Nao
Select your robot in the menu and make sure that Raw Vision is ticked.
Click record on the bottom. You should now see some vision information and images in the vision tab.
Run the robot around or whatever you want it to see.
Save the ofn by going File -> Save File.

Go to the colour calibration tab. You should see an image there.
You can set what colours are shown in the overlay. (Selected is the colour selected in the menu above this)
With the selected colour start clicking the regions you want to assign. It should assign that colour and other colours similar to it.
If you want the selection to not pick up as many colours you can select Auto-weight gaussians.
To save the file click Save. This should no generate the colour information, which would be pushed to the robot on the next nao_sync.
You can view the updated saliency image in the vision tab by selecting Vision -> Reload Both NNMCs

If you don't want it to save the calibration kernel to the default location click Save As...
If you want to start again in the calibration select New on the camera that you want to calibrate and do the same as above.
There are two calibrations for each of the cameras, changes to one will not affect the other. Also when you click save it only saves the current camera, so you should change camera and click save on the other if you made changes to that as well. You can save them both by going Calibration -> Save both kernels
You can load a kernel which you have saved at another time by clicking load. It loads the kernel into the currently selected camera.

Differentiate field and background
Green should be underclassified so that there is almost no green above the field, but the field should be mostly green
Background on the field is ok, provided its mainly green

Orange should be classified so that there is none on any waistbands, none meaning absolutely zero
The ball needs to have some orange on it, but doesn't need to be entirely orange
In particular, dont worry about the top and bottom of the ball which are often significantly lighter/darker respectively
Its fine for the ball to have robot red in it

Should be underclassified
Just make sure the wasteband has some blue on it and that the field isn't covered in blue

This should be easy
Just make sure there isnt any on the goal posts
Also be careful that you classify all the field lines, the overlay white is a similar colour to the actual line, so sometimes it can look classified
To get around this, turn on 'all colours + unclassified'
Dont be surprised if lots of things are white (in particular robots)

Let Sean ([email protected]) know if anything doesnt make sense and apologies for any typos.