Implement capture method on Vision.py

[avesk]

This feature depends on https://github.com/uvic-aero/onboard-computer/issues/38

Overview

Vision.py will capture images with the piCam in order to algorithmically determine a vector for the vehicle's trajectory.

Instructions:

The logic for image capturing will live within apps/Vision/Capture.py.

Write a class method called capture() which accepts a boolean value, use_picam, for whether to use PiCam or the default camera (your webcam).

The method should conditionally use cv2.VideoCapture(0) or PiCam.take_picture depending on the value of use_picam.

• Look at https://github.com/uvic-aero/workshops/blob/master/video-stream/node-udp/python-server.py for details on how to implement opencv's image capturing library to read frames from your webcam.
• You'll need to import the PiCam singleton from https://github.com/uvic-aero/onboard-computer/blob/master/apps/PiCam/piCam.py in order to have access to its methods.

Testing

Try and test the opencv image capturing with your webcam.

Don't worry about testing PiCam.take_picture yet. If you're keen, you'll need to test your branch on a Raspberry Pi with a connected PiCam peripheral.