Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/alvinwan/motiosecure
toy security platform with motion detection implementation for cheap, low-quality camera lens
https://github.com/alvinwan/motiosecure
desktop ios mac mobile motion-detection python security swift
Last synced: 29 days ago
JSON representation
toy security platform with motion detection implementation for cheap, low-quality camera lens
- Host: GitHub
- URL: https://github.com/alvinwan/motiosecure
- Owner: alvinwan
- License: apache-2.0
- Created: 2017-06-16T13:06:12.000Z (over 7 years ago)
- Default Branch: master
- Last Pushed: 2017-07-06T06:26:13.000Z (over 7 years ago)
- Last Synced: 2024-11-12T17:50:25.987Z (3 months ago)
- Topics: desktop, ios, mac, mobile, motion-detection, python, security, swift
- Language: CSS
- Homepage:
- Size: 846 KB
- Stars: 3
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# MotioSecure
Simple security platform with motion detection implementation for cheap, low-quality camera lens
**Mac and iOS apps are coming soon**
Ever have a friend hijack your computer while you're away? Ever wanted to catch them red-handed? Run this application to do just that. With just an iOS app and a mac app, you'll be good to go. **Applications are pending review. Want it now? See Development installation instructions below.**
- Sends a notification to your phone when motion is detected
- Saves all detected motion as mp4 videos for playback
- Surprise the intruder with a live camera feed, and engage with your invader in conversation!
- Uses Apple security protocols to make live feed, notifications available; no janky, inter-device, unencrypted communication
> This is not a serious security platform. For all intents and purposes, I wrote it to prank my friends. With that said, the application actually does what I claim above. I make no guarantees about its ability to prevent theft or its security. Please don't use this seriously.
# How it Works
On server-side, `opencv` abstracts away the camera. We compute the difference across `l` timesteps and use `scipy`'s `SVD` to analyze explained variance ratios. If the sum of the top `k` ratios exceed a threshold, we consider motion to be detected.
`opencv` provides additional utilities that allow us to encode and write to mp4 videos. A Python interface for Apple's Push Notification Service `pyapns` allows the Python app to interface with iOS notifications accordingly. A third-party library `websockets` is used to launch a local socket, giving the application live updates for "motion detected" or not.
- See the [`web/` README](http://github.com/alvinwan/motiosecure/tree/master/web) for more on a locally-hosted web application.
- See the [`desktop/` README](http://github.com/alvinwan/motiosecure/tree/master/desktop) for how we packaged the application for Mac App Store distribution.
- See the [`ios/` README](http://github.com/alvinwan/motiosecure/tree/master/ios) for the mobile component.
