Facedetection_HaarCascade

Face Detection using openCV with HaarCascade

Description

This project demonstrates a simple face detection application using OpenCV and Python. It captures video from the webcam, detects faces in real-time, and draws rectangles around detected faces.

Screenshots

Demo Video

You can view a demo of the face detection application here.

https://github.com/Syed-Basila/Facedetection_HaarCascade/assets/123718024/a084a6ae-6f3b-467b-90e5-95f46c26f111

Features

• Real-time face detection
• Drawing rectangles around detected faces

  Installation

  Prerequisites

• Python 3.x
• OpenCV

Install required packages:

pip install opencv-python

Download Haar Cascade file:

Download the haarcascade_frontalface_default.xml file from OpenCV GitHub repository and place it in the project directory.

Usage

Run the following command to start the face detection application:

python face_detection.py

Face Detection Script

import cv2 # openCV

alg= "haarcascade_frontalface_default.xml" #accessed the model file
haar_cascade=cv2.CascadeClassifier(alg) #loading the model with cv2

cam = cv2.VideoCapture(0) #intializing camera

while True:
    _,img = cam.read() #read the frame from the camera
    grayImg = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY) #converting color into gray scale
    face = haar_cascade.detectMultiScale(grayImg,1.3,4) #get coordinates of face
    for (x,y,w,h) in face: #segregating x,y,w,h.
        cv2.rectangle(img,(x,y),(x+w, y+h), (0,255,0),2)
    cv2.imshow("FaceDetection",img)
    key = cv2.waitKey(10)
    if key ==27:#esc button to return
        break

cam.release()
cv2.destroyAllWindows()

output

https://syed-basila.github.io/Facedetection_HaarCascade/

Haar Cascades

Haar Cascades are a popular object detection method used in computer vision, developed by Paul Viola and Michael Jones. They are especially known for their use in real-time face detection. The technique involves training a cascade function with a large number of positive and negative images. The trained model can then detect objects in new images. In this project, we use the haarcascade_frontalface_default.xml file, which is pre-trained to detect faces.

Contributing

Contributions are welcome! Please create an issue to discuss any changes or improvements.