Real Time Face Identification, Creating DataSet and Train Images

Introduction

This tutorial will guide you through the process of implementing real-time human face recognition using Python and OpenCV. We will create a dataset of images, train a model using the Haar Cascade Classifier, and ultimately achieve face identification in real-time. This beginner-friendly project will enhance your understanding of computer vision and machine learning concepts.

Step 1: Set Up Your Environment

Before diving into code, ensure you have the necessary software and libraries installed.

• Install Python (version 3.6 or higher).
• Install OpenCV and other required libraries. You can do this using pip:

  pip install opencv-python opencv-python-headless
  

• Clone the GitHub repository containing the project files:

  git clone https://github.com/yashwantpathakrjit/Real-Time-Face-Identification-Creating-DataSet-and-Train-Images.git
  

Step 2: Create a Dataset

To train your face recognition model, you need a dataset of images.

• Import the necessary libraries:

  import cv2
  import os
  

• Create a directory to store the images:

  os.makedirs('dataset', exist_ok=True)
  

• Use the following code snippet to capture images:

  def capture_images(name):
      cam = cv2.VideoCapture(0)
      detector = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
      count = 0
      while True:
          ret, frame = cam.read()
          gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
          faces = detector.detectMultiScale(gray, 1.3, 5)
          for (x, y, w, h) in faces:
              cv2.rectangle(frame, (x, y), (x + w, y + h), (255, 0, 0), 2)
              count += 1
              cv2.imwrite(f"dataset/{name}.{count}.jpg", gray[y:y + h, x:x + w])
          cv2.imshow('Capturing Images', frame)
          if cv2.waitKey(1) & 0xFF == ord('q') or count >= 100:
              break
      cam.release()
      cv2.destroyAllWindows()
  

• Call the function to start capturing images for a specific individual:

  capture_images("person_name")
  

Step 3: Train the Face Recognition Model

Once the dataset is ready, you can train the model.

• Load the images and labels:

  def load_images_and_labels():
      images = []
      labels = []
      for root, dirs, files in os.walk('dataset'):
          for file in files:
              img = cv2.imread(os.path.join(root, file), cv2.IMREAD_GRAYSCALE)
              images.append(img)
              labels.append(int(file.split('.')[0]))  # Assuming labels are integers
      return images, labels
  

• Use the EigenFaceRecognizer or FisherFaceRecognizer to train:

  recognizer = cv2.face.LBPHFaceRecognizer_create()
  images, labels = load_images_and_labels()
  recognizer.train(images, np.array(labels))
  recognizer.save('trainer.yml')
  

Step 4: Implement Real-Time Face Recognition

Finally, set up the code for real-time face recognition.

• Load the trained model and Haar Cascade:

  recognizer.read('trainer.yml')
  detector = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
  

• Capture video from the webcam and recognize faces:

  video_capture = cv2.VideoCapture(0)
  while True:
      ret, frame = video_capture.read()
      gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
      faces = detector.detectMultiScale(gray, 1.3, 5)
      for (x, y, w, h) in faces:
          id_, conf = recognizer.predict(gray[y:y + h, x:x + w])
          if conf >= 45:  # Confidence threshold
              cv2.putText(frame, str(id_), (x, y - 10), cv2.FONT_HERSHEY_SIMPLEX, 1, (255, 0, 0), 2)
          cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 2)
      cv2.imshow('Real-Time Face Recognition', frame)
      if cv2.waitKey(1) & 0xFF == ord('q'):
          break
  video_capture.release()
  cv2.destroyAllWindows()
  

Conclusion

In this tutorial, you learned how to create a real-time face identification system using Python and OpenCV. You captured images to create a dataset, trained a face recognition model using the Haar Cascade Classifier, and implemented real-time detection.

Next steps could include exploring other algorithms for face recognition, enhancing the dataset with more images, or integrating this system into a larger application. Happy coding!