OpenCV: Detecting and Tracking Faces

Breaking Down the Code

Import the Libraries

from adafruit_servokit import ServoKit 
import cv2 
import time

Create the ServoKit Object.

kit=ServoKit(channels=16)

Create Pan and Tilt Objects and Apply Them to the ServoKit Object.

pan=0
tilt=135

kit.servo[0].angle=pan
kit.servo[1].angle=tilt

Load the Face Cascade Classifier:

OpenCV provides pre-trained cascade classifiers for detecting faces. You can download the XML file for the face cascade classifier from the OpenCV GitHub repository. Once downloaded, load the classifier using the cv2.CascadeClassifier function.

face_cascade = cv2.CascadeClassifier('path_to_xml_file/haarcascade_frontalface_default.xml')

Initialize the Video Capture:

Use the cv2.VideoCapture function to initialize the video capture. You can pass the argument 0 to capture video from the default camera or specify the path to a video file.

cap = cv2.VideoCapture(0)

Read and Process Frames:

Continuously read frames from the video capture and process them for face tracking. Convert the frames to grayscale for better face detection performance.

while True:
    ret, frame = cap.read()
    if not ret:
        break

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

Detect Faces in the Frame:

Use the detectMultiScale function of the face cascade classifier to detect faces in the grayscale frame.

faces = face_cascade.detectMultiScale(gray, scaleFactor=1.1, minNeighbors=5, minSize=(30, 30))

Adding Bounding Box to Faces:

Iterate over the detected faces and draw rectangles around them on the frame.

for (x, y, w, h) in faces:
    cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 2)

Track the Faces:

Create conditions using mathematical operations to calculate the Pan and Tilt servo positions.

Xcent=x+w/2
Ycent=y+h/2
errorPan=Xcent-width/2
errorTilt=Ycent-height/2

if abs(errorPan)>15:
    pan=pan-errorPan/30
if abs(errorTilt)>15:
    tilt=tilt+errorTilt/30

Add Error Correction:

By adding range limitations, we can fix out of range errors.

if pan>180:
    pan=180
    print('Pan out of range')
if pan<0: pan=0 
    print('Pan out of range') 
if tilt>180:
    tilt=180
  print('Tilt out of range')
if tilt<0:
    tilt=0
    print('Tilt out of range')

Control servo positions:

By applying the updated Pan and Tilt positions each time through the loop.

kit.servo[0].angle=pan 
kit.servo[1].angle=tilt
roi_gray=gray[y:y+h, x:x+w]
roi_color=frame[y:y+h, x:x+w]

Display the Frame:

Show the processed frame with the tracked faces using the cv2.imshow function.

cv2.imshow('Face Tracking', frame)

Exit the Program:

Add a condition to exit the loop and release the video capture when a specific key (e.g., ‘q’) is pressed.

if cv2.waitKey(1) == ord('q'):
    break

cap.release()
cv2.destroyAllWindows()