tl;dr

Docker image for Raspberry Pi Camera (i.e. picam) libraries and utilities to run the picam for different applications in Docker.

Utility Scripts

What follows are a collection of simple scripts (i.e. python, and shell scripts) to help with different picam-related tasks.

timelapse.sh

Just a simple shell script to wrap all the timelapse-related features. The script will simply capture a photo every minute for 12 hours, and annotate the datetime onto the center of the picture (for aiding analysis on later viewings). The script can be run as follows. First download the repo:

$ git clone https://github.com/RagingTiger/docker-picam

Then cd to the src/shell directory:

$ cd docker-picam/src/shell`

And finally run the script (in the background) as follows:

$ bash timelapse.sh &

Post-Processing

Here we will discuss various post-processing methods and tools.

timelapse video

After we have recorded our timelapse images we can "stitch" them together using ffmpeg.¹ We can use the tigerj/hckr docker image to do this (Note: this can be an I/O and CPU intensive process so it is recommended that you run these commands on the most powerful machine you have). First cd into the directory with all the timelapse images:

$ cd /my/timelapse/images/

Then run a container of the tigerj/hckr image:

$ docker run \
           --rm \
           -v /my/timelapse/image:/home/hckr \
           -it \
           tigerj/hckr

This will launch an interactive bash prompt (running in the docker container) where we can run the ffmpeg commands:

# ffmpeg -r 12 -pattern_type glob -i '*.jpg' -c:v copy output.avi

This command will simply stitch the pictures together at a rate of 12 frames per second (fps). But the resulting output.avi will be rather large, so it would be good to reduce that size. We can do that, again using ffmpeg, as follows:

# ffmpeg -i output.avi -c:v libx264 -preset slow -crf 15 output-final.mkv

The resulting mkv file (output-final.mkv) will be significantly smaller, but can be compressed even more by using a tool known as handbrake of which an excellent docker image exists created by jlesage at jlesage/handbrake. Click the previous links to learn more about handbrake and the jlesage/handbrake image.

References

Recording a timelapse with a Raspberry Pi ↩