Making Raspberry Pi in the NoIR - Part IX - field testing
Not sure what this blog is about? Go back to Part I to find out.
To fully test your camera, you will want to set up the router so that you can connect to the camera trap (the camera unit in the container) and tweak any necessary settings using your laptop and the Pikrellcam web interface. Connect your laptop to the router network, open a web browser and navigate to Pikrellcam using:
Note: you will need to change the IP address to suit your camera trap.
With your powered Pi you will see the camera screen in the web browser. You can test your lights using the scripts you installed earlier. Remember, if you want to check the NoIR set up, you will need to go somewhere dark and turn on the IR lights.
> ssh firstname.lastname@example.org
> cd /home/pi/brightpi-control/
You may notice that the Pi cameras are not great when it comes to fine focus. You can improve this a bit by turning the camera lens anti-clockwise. I used some fine forceps to grasp the camera lens and twist it. Twist a little bit at a time - you don't want to break it or have it fall off.
This video (above) shows a test of the NoIR camera set up with motion detection. I was lucky enough to have this Phalaenopsis and fly as my test system. You can see that the motion detection works quite well. The fine focus is a little off. The camera unit is placed about 20 cm away from the flowers. Note: the colour is pinkish because it it the NoIR camera being used during the day.
This video (above) is still using the NoIR camera but I adjusted the fine focus a little by twisting the camera lens anti-clockwise. The camera unit is placed about 20 cm away from the flowers.
This video (above) shows the difference between running IR and white lights. I had not changed the fine focus at all. Alas, I could not convince the fly to stay in the dark cupboard with me so I used the fine forceps to trigger motion. The camera unit is placed about 35 cm away from the flowers. As you can see, the IR lights use is not providing a crystal clear image, but I don't think you should expect them to. I do think the IR light clarity can be improved with some fine focus adjustment. I also noted that 'white' is not the best colour to try and detect via IR - various tests on food labels show that white text, for example, is not always clear.
The white LEDs improve the image greatly, but they come with the caveat that they may modify nocturnal insect behaviour, so I would rather avoid using them. I'll post an update on the performance once I run some more field tests.
How do I access my saved video files?
Your video files, thumbnails and stills will save to the following locations on your Pi unit:
You can login to the Pi unit wirelessly and transfer all files at once to your laptop using:
> scp email@example.com:/home/pi/pikrellcam/media/videos/*.mp4 /Users/yourname/Documents/FileForPiStuff
> scp firstname.lastname@example.org:/home/pi/pikrellcam/media/stills/*.jpg /Users/yourname/Documents/FileForPiStuff
If you want to delete all the thumbnails, for example, you can:
> ssh email@example.com
> cd /home/pi/pikrellcam/media/thumbnails
> rm -r *.jpg