A great way to capture awesome sunset photos (and other iteresting scenes)!

The capturer runs as a daemon and uses the Pi camera to capture images every few seconds, compares that new image to the last one that was saved, and saves the new one if it is "interesting" enough. Interesting is determined by the how different the new image is from that last saved one, by measuring entropy.

You can adjust the entropy threshold to your environment, which results in more or less images being captured with fewer or greater differences. Please take a look at the beginning of the source code for 'THRESHOLD'. A higher value means the images are more different before saving a file.

The capturer generates an index.html page that shows that most recent (in the past 24 hours) images captured. You will need to install a web server and configure the 'INDEX_HTML' directory constant in the capturer.py file to make this work.

Installation

• Setup your Pi with the latest Raspian and Pi Cam
• Execute: sudo apt-get install python-scipy -y
• Execute: sudo apt-get install python-picamera -y
• Install Apache: sudo apt-get install apache2 -y
• Install gdata: See https://developers.google.com/gdata/articles/python_client_lib?csw=1#linuxZ
• Create the directory '/home/pi/capturer'
• Execute: mkdir /home/pi/capturer/latest
• Create the directory '/home/pi/capturer/images' assuming /home/pi/capturer is where you want it.
• Execute: sudo ln -s /home/pi/capturer/latest /var/www/latest
• Execute: sudo ln -s /home/pi/capturer/images /var/www/latest/images
• Execute: sudo ln -s /home/pi/capturer/images /var/www/images
• Copy capturer.init to /etc/init.d
• Copy capturer.py to /home/pi/capturer
• Copy settings.py to /home/pi/capturer
• Copy __init__.py to /home/pi/capturer
• Review settings.py and make any required updates
• Execute: sudo update-rc.d capturer defaults
• Restart your Raspberry Pi or execute: sudo service capturer start
• Execute: 'tail -f /var/log/capturer.log' and watch the log for errors
• Type 'CTL-C' to break out of the log tailing
• You should now have have interesting photos stored in the images directory

How it works

• Look for CTL-C presses, in case this is run interactively
• Initializes logging and remaps stderr and stdout to logs
• Capture the first image and save it as a jpg file in the images directory
• Capture the second image
• Compute the entropy of the currently saved image and the new one
• Save the new image, if the entropy is greater than or equal to the threshold
• Repeat this process forever