movement-v2.py

import cv2
import keyboard
import time

color = {"blue":(255,0,0), "red":(0,0,255), "green":(0,255,0), "white":(255,255,255)}

# Method to detect nose
def detect_nose(img, faceCascade):
    
     # convert image to gray-scale
    gray_img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    
    # detecting features in gray-scale image, returns coordinates, width and height of features
    features = faceCascade.detectMultiScale(gray_img, 1.1, 8)
    nose_cords = []
    # drawing rectangle around the feature and labeling it
    for (x, y, w, h) in features:
#         cv2.rectangle(img, (x,y), (x+w, y+h),  color['green'], 2)  #uncomment if you want to see face boundary
        cv2.circle(img, ((2*x+w)//2,(2*y+h)//2), 10, color['green'], 2) 
        nose_cords = ((2*x+w)//2,(2*y+h)//2)
    return img, nose_cords

def draw_controller(img, cords):
    size = 40
    x1 = cords[0] - size
    y1 = cords[1] - size
    x2 = cords[0] + size
    y2 = cords[1] + size
    cv2.circle(img, cords,  size, color['blue'], 2) 
    return [(x1,y1), (x2,y2)]

# tracking sudden changes
def get_movement(curr, prev, last_time_update, cmd):
    if not (len(curr) > 0 and len(prev) > 0):
        return last_time_update, cmd
    xc, yc = nose_cords
    tc = time.time()
    ox,oy, to = prev_cords
    diffx = xc - ox
    diffy = yc - oy
    thres_diff = 50
    
    thres_diff_t = 1
    if last_time_update + 0.4 > tc:
        return last_time_update, cmd
        
    if (abs(diffx)>thres_diff or abs(diffy) > thres_diff) and abs(tc-to)<thres_diff_t:
        if abs(diffx) > abs(diffy):
            if diffx > 0:
                cmd = "right"
            else:
                cmd = "left"
        else:
            if diffy > 0:
                cmd = "down"
            else:
                cmd = "up"
        print("Movement detected: ", cmd, "\n")
        keyboard.press_and_release(cmd)
        last_time_update = time.time()
    return last_time_update, cmd
    
# Loading classifiers
faceCascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')

# Capturing real time video stream.
video_capture = cv2.VideoCapture(-1)

# get vcap property 
width  = video_capture.get(3) # float
height = video_capture.get(4) # float
cmd = ""
prev_cords = (0,0,  time.time())
last_time_update = time.time()

    
while True:
    # Reading image from video stream
    _, img = video_capture.read()
    img = cv2.flip( img, 1 )

    # detect nose and draw
    img, nose_cords = detect_nose(img, faceCascade)
    cv2.putText(img, cmd, (10,50), cv2.FONT_HERSHEY_SIMPLEX, 1, color['red'], 1, cv2.LINE_AA)
    
    last_time_update, cmd = get_movement(nose_cords, prev_cords, last_time_update, cmd)
    
    x, y = (0,0) if len(nose_cords) == 0 else nose_cords
    prev_cords = (x,y,time.time())

    # Writing processed image in a new window
    cv2.imshow("face detection", img)
    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

# releasing web-cam
video_capture.release()
# Destroying output window
cv2.destroyAllWindows()