The classes found in this package allow users to run clock-time based timelapses using Node.JS, gPhoto2, and optionally SunCalc & AWS. This project was designed specifically to run headless on a Raspberry Pi, using Forever.

Sample Videos

Check out this YouTube Channel to view videos created using this codebase.

Basic Installation

This package relies heavily on external software. Below you'll find a crash course to get your Pi up and running. Most other machines should come up by following similar steps.

1. Install gPhoto2 using the gPhoto2 Updater (https://github.com/gonzalo/gphoto2-updater)

$ wget https://raw.githubusercontent.com/gonzalo/gphoto2-updater/master/gphoto2-updater.sh && chmod +x gphoto2-updater.sh && sudo ./gphoto2-updater.sh

2. Install Node.JS. Note that on the Pi, we must use an ARM distribution (http://node-arm.herokuapp.com/) and for node-gphoto2, we must be using Node 10...

wget http://node-arm.herokuapp.com/node_0.10.35_armhf.deb
sudo dpkg -i node_0.10.35_armhf.deb

Non Raspberry Pi users may want to consider using NVM to maintain multiple versions of Node. More information can be found at https://github.com/creationix/nvm

3. Clone this repo...

$ git clone git@github.com:brettcrowell/node-gphoto2-timelapse.git

4. Install Node dependencies...

$ cd node-gphoto2-timelapse
$ npm install

USB Reset Script

During the course of a long-running capture, gPhoto2 may lose its connection to the camera. If this situation is detected, node-gphoto2-timelapse will execute a script which resets the USB port the camera is connected to.

In order for this all to work, that script (usbreset.c) must be compiled using gcc before running the app...

$ gcc -o usbreset usbreset.c

Usage

As mentioned above, this package allows you to create timelapses based on clock-time. This is facilitated by two main classes, Sequence and Timelapse.

A Sequence is a specialized queue whose elements are exposure times, represented as Objects containing UTC timestamps and metadata such as a filename prefix and S3 Bucket (if applicable). A sample object follows...

{
  name: 'sunrise',
  bucket: 'bcrowell-timelapse',
  ts: 1425380400000
}

Note: If bucket is omitted, no attempt will be made to upload to S3, and images will be stored in the ./output directory

Sequence exposes two methods to its consumer, getNextImage and hasMoreImages. Unlike a queue, getNextImage won't necessarialy return the object on top of the stack, but rather the next image whose timestamp is in the future. If a sequence contains many exposures with timestamps that have passed, it will skip them. This becomes extremely helpful in situations where a long-running capture might be interrupted.

When a Timelapse is created, a Sequence must be passed in. Timelapse will then handle connecting to the camera, capturing each image at the correct time, downloading it from the camera, and saving or uploading to Amazon S3.

var tl = require('../timelapse.js');
var seq = require('../sequence.js');

var now = new Date().getTime();
var myTestSequence = new seq.Sequence();

// take 30 exposures, one every 30 seconds, starting immediatey

for(var i = 0; i < 30; i++){
  myTestSequence.addImage({
    name: 'test-lapse',
    ts: now + (i * 30000)
  });
}

new tl.Timelapse(myTestSequence);

To run the example above, assuming it is in a file called simple-sample.js, simply point node there...

$ node ./samples/simple-sample.js

For longer running lapses, it can be very useful to install and use forever...

$ npm install -g forever
$ forever ./samples/simple-sample.js

More information on forever can be found at https://github.com/foreverjs/forever

For a more complicated lapse based on sun position, check out solar-sample.js

Tips and Tricks

• On Ubuntu, cameras must be unmounted in order to be reserved for node-gphoto2-timelapse

License

MIT © Brett Crowell