I have a webcam taking a photo every minute. Most of the time, there is nothing happening, so I end up with thousands of thousands of near identical photos. How can I find the photos where something actually happened?

Note the lighting may vary somewhat between photos. This should be ignored, I only want to find the photos where something actually changed or moved. These 'interesting' photos should be highlighted in some way, maybe copied to a different folder. And then some way of filtering out or bulk deleting the rest of the photos.

The software should run on Windows 7, and free of charge. Bonus if it is also free/open source.

1 Answer 1


You can do the sort of thing that you are looking for with the combination of the Open Computer Vision libraries and python including setting a threshold for the level of difference, (due to lighting, etc.), that you discard the image as being no different to the previous.

There is an nice write-up on motion detection part 1 & part 2 that is aimed at detecting significant changes using a RaspberryPi and live web cam but you should be able to adapt it quite easily for your use as it is very well explained what is being done and why.

  • Free, gratis & open source
  • Cross Platform: Windows 10, and 7 and OS-X and Linux, etc.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.