Reading camera frames

This first example, read_camera.py, shows you how to read frames from a camera that's connected to your computer. The required argument is index_camera, which indicates the index of the camera to read. If you have connected a webcam to your computer, it has an index of 0. Additionally, if you have a second camera, you can select it by passing 1. As you can see, the type of this parameter is int. 

The first step to work with cv2.VideoCapture is to create an object to work with. In this case, the object is capture, and we call the constructor like this:

 # We create a VideoCapture object to read from the camera (pass 0):
capture = cv2.VideoCapture(args.index_camera)

If index_camera is 0 (your first connected camera), it is equivalent to cv2.VideoCapture(0). To check whether the connection has been established correctly, we have the capture.isOpened() method, which returns False if the connection could not be established. In the same way, if the capture was initialized correctly, this method returns True.

To capture footage frame by frame from the camera, we call the capture.read() method, which returns the frame from the camera. This frame has the same structure as an image in OpenCV, so we can work with it in the same way. For example, to convert the frame into grayscale, do the following:

gray_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

Additionally, capture.read() returns a bool. This bool indicates whether the frame has been correctly read from the capture object.