(Feb. 12-18) "Hands" on Experience

This week in shop, I used what I had learned about OpenCV and computer vision in general to program hand detection software with a pre-trained model that I found on Github. Using a the python plugins: Tensorflow, Mediapipe, and OpenCV, I was able to track the position of key points on the hand. I wanted to use this new software to control a robotic hand, but learning kinematics in less than three months might be difficult. I will most likely take Mr. Christy's advice and try to use the gestures as a controller for a variety of things including a robotic arm.

while True:
    # Read each frame from the webcam
    _, frame = cap.read()

    x, y, c = frame.shape

    # Flip the frame vertically
    frame = cv2.flip(frame, 1)
    framergb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)

    # Get hand landmark prediction
    result = hands.process(framergb)

    # print(result)

    className = ''

    # post process the result
    if result.multi_hand_landmarks:
        landmarks = []
        for handslms in result.multi_hand_landmarks:
            for lm in handslms.landmark:
                #print(id, lm)
                lmx = int(lm.x * x)
                lmy = int(lm.y * y)

                landmarks.append([lmx, lmy])

            # Drawing landmarks on frames
            mpDraw.draw_landmarks(frame, handslms, mpHands.HAND_CONNECTIONS)