@bytesnz

Jack Farley, Computer Engineer

The next place I decided to target for my litter picker antics was the paths along Hoe Stream and the White Rose Lane Local Nature Reserve (OSM). Having fixed the issues with the GPS positions on the Pi and the other issues I found, I was hopeful of great success with the litter picker.

As I was walking to the park I found that the GPS module was taking a very very long time to get a good fix. I pulled out my phone and fired up the track recording in OSMAnd so I could at least track where I went, which magically made the GPS module on the Pi get a good fix.

Missing the bridge as I was paying too much attention to my phone and the muddy puddles, I started from Hoe Bridge School Sports Grounds and worked back to the nature reserve.

On previous walks, I had seen a couple of bits of trash caught by a tree that had fallen in to the Hoe Stream, so once I had gone round the nature reserve once, headed to the tree. It was a lot worse than I thought. I picked out a dozen bottles, four tennis balls, a nerf dart and a strange black floating ball, which shattered when I tried to crush it (still not sure what it is) and found I was out of space in my bag - will go back with a drysuit to get the rest.

On returning home, I found again that it hadn't recorded positions. It turned out it was due to some wires shorting the serial lines on the GPS modules that I hadn't cleaned up after soldering them (bad Jack). Since I had recorded my track on my phone, I was able to add GPS positions to the images by correlating the time the photo was taken with the position data from track using Digikam (select the images, open the "Item" menu, select "Edit Geolocation", change to the "GPS Correlator" tab, load the GPX file, choose the correct timezone, click "Correlate" and voilà).

The tally for the picking was:

• 3 face masks
• 1 medical glove
• 4 tennis balls
• 3 bags of dog poop
• 2 empty dog poop bags
• 3 wet wipes
• 19 single use plastic bottles
• 1 plastic drink bottle
• 3 glass bottles
• 15 cans
• 1 pair of garden gloves
• 2 takeaway containers
• 1 juice pouch
• 1 yoghurt container
• 11 snack wrappers
• a couple of other bits of plastic bag

As before here is a video of the images from the pi-trash-cam.

Here is the commands I used to create the video

$ # Resize the images
$ for f in `ls *.jpg`; do convert -auto-orient -resize '640x480^' -quality 80 "$f" "resized/$f"; done
$ # Create a video from the images
$ convert -delay 10 resized/* trash.mpg
$ # Encode as 
$ ffmpeg  -i trash.mpg -vcodec libx264 trash.mp4