Introduction

Adrian Rosebrock posted a blog , named ” Home surveillance and motion detection with the Raspberry Pi, Python, OpenCV, and Dropbox“. This project is very cool, it can take a picture if there are things moving , then upload to Dropbox.

I can’t help thinking about how to port this project onto pcDuino8 Uno. Yes, this is not very hard!

Pre-requisites:

Hardware

• pcDuino8 Uno
• USB UVC camera(also called as webcam)

Software

• Ubuntu 14.04
• Python 2.7.6
• OpenCV 2.4.11
• imutils

Steps

1. Download source code

Based on the code that Adrian Rosebrock shared, I remove the function that can upload the pictures to Dropbox. Instead, I save these pictures to the local. You can get this source code from github.

$ git clone https://github.com/YaoQ/motion-detection-with-opencv

2. Connect webcam

Plug the webcam input pcDuino8 Uno, and power on. the webcam could be recognized as /dev/video0.

3. Test

$ python motion-detector.py -c conf.json

If you use different types of webcams or want to change the interval time to save a picture, and so on, you should modify these parameters in conf.json.