First commit

Author: demoncoder-crypto

app.py

@@ -0,0 +1,104 @@
+from flask import Flask, Response, render_template, jsonify, request
+import cv2
+import numpy as np
+
+app = Flask(__name__)
+
+
+recognizer = cv2.face.LBPHFaceRecognizer_create()
+recognizer.read('trainer.yml')
+face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
+eye_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_eye.xml')
+
+
+cap = cv2.VideoCapture(0)
+
+
+color_recognized = (0, 255, 0)  
+color_unrecognized = (0, 0, 255)  
+font = cv2.FONT_HERSHEY_SIMPLEX
+confidence_threshold = 100  
+
+def preprocess_face(face):
+    """Preprocess the face image to improve recognition."""
+    face = cv2.equalizeHist(face)  
+    face = cv2.GaussianBlur(face, (3, 3), 0)  
+    face = cv2.resize(face, (200, 200))  
+    return face
+
+def align_face(face):
+    """Align the face image to improve recognition."""
+    return face
+
+def gen_frames():
+    """Generate frames from the webcam."""
+    while True:
+        ret, frame = cap.read()
+        if not ret:
+            break
+
+        gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
+
+        
+        faces = face_cascade.detectMultiScale(gray, scaleFactor=1.1, minNeighbors=5, minSize=(30, 30))
+
+        for (x, y, w, h) in faces:
+            face = gray[y:y+h, x:x+w]
+            preprocessed_face = preprocess_face(face)
+            aligned_face = align_face(preprocessed_face)
+            label, confidence = recognizer.predict(aligned_face)
+
+           
+            eyes = eye_cascade.detectMultiScale(face)
+
+            
+            if confidence < confidence_threshold and len(eyes) >= 2:
+                color = color_recognized
+                text = 'Welcome Sir'
+            else:
+                color = color_unrecognized
+                text = 'Cannot enter'
+
+            
+            cv2.rectangle(frame, (x, y), (x+w, y+h), color, 2)
+            cv2.putText(frame, text, (x, y-10), font, 0.9, color, 2, cv2.LINE_AA)
+
+        
+        ret, buffer = cv2.imencode('.jpg', frame)
+        frame = buffer.tobytes()
+        yield (b'--frame\r\n'
+               b'Content-Type: image/jpeg\r\n\r\n' + frame + b'\r\n')
+
+@app.route('/')
+def index():
+    return render_template('index.html')
+
+@app.route('/video_feed')
+def video_feed():
+    return Response(gen_frames(),
+                    mimetype='multipart/x-mixed-replace; boundary=frame')
+
+@app.route('/status', methods=['GET'])
+def status():
+    
+    return jsonify({
+        'status': 'Authentication in progress',
+        'progress': '50%'
+    })
+
+@app.route('/start_auth', methods=['POST'])
+def start_auth():
+    
+    return jsonify({'status': 'Authentication started'})
+
+@app.route('/stop_auth', methods=['POST'])
+def stop_auth():
+    
+    return jsonify({'status': 'Authentication stopped'})
+
+if __name__ == '__main__':
+    try:
+        app.run(debug=True)
+    finally:
+        cap.release()
+        cv2.destroyAllWindows()

dataset/Screenshot 2024-08-25 0049077 - Copy (10).png

@@ -0,0 +1,60 @@
+import cv2
+import os
+
+
+face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
+eye_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_eye.xml')
+
+
+output_dir = 'dataset'
+if not os.path.exists(output_dir):
+    os.makedirs(output_dir)
+
+
+cap = cv2.VideoCapture(0)
+
+print("Collecting images. Press 'q' to quit.")
+
+count = 0
+user_id = input("Enter user ID: ")
+
+while True:
+    ret, frame = cap.read()
+    if not ret:
+        break
+
+    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
+    
+    
+    faces = face_cascade.detectMultiScale(gray, scaleFactor=1.1, minNeighbors=5, minSize=(30, 30))
+
+    for (x, y, w, h) in faces:
+        face = gray[y:y+h, x:x+w]
+        color_face = frame[y:y+h, x:x+w]
+        
+        
+        eyes = eye_cascade.detectMultiScale(face)
+        
+        
+        cv2.rectangle(frame, (x, y), (x+w, y+h), (255, 0, 0), 2)
+        
+        
+        for (ex, ey, ew, eh) in eyes:
+            cv2.rectangle(color_face, (ex, ey), (ex+ew, ey+eh), (0, 255, 0), 2)
+        
+        
+        if len(eyes) >= 2:  
+            face_filename = os.path.join(output_dir, f'user.{user_id}.{count}.jpg')
+            cv2.imwrite(face_filename, face)
+            print(f"Image {count} saved.")
+            count += 1
+
+    
+    cv2.imshow('Collecting Images', frame)
+
+    
+    if cv2.waitKey(1) & 0xFF == ord('q') or count >= 10:
+        break
+
+cap.release()
+cv2.destroyAllWindows()