Update main.py

main.py

@@ -30,10 +30,9 @@ def load_student_encodings(folder_path: str):
     if not os.path.exists(folder_path):
         return {"error": f"Folder '{folder_path}' does not exist"}
 
-    # Create face detector and feature extractor
+    # Create face detector
     face_detector = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
-    face_recognizer = cv2.face.LBPHFaceRecognizer_create()
-
+    
     # Check if a model already exists for this folder
     model_path = os.path.join(folder_path, "face_model.pkl")
     if os.path.exists(model_path):
@@ -73,38 +72,41 @@ def load_student_encodings(folder_path: str):
 
 def process_image(image_array, students):
     """Compare uploaded image with stored student images using OpenCV."""
-    # Convert to grayscale
-    gray = cv2.cvtColor(image_array, cv2.COLOR_BGR2GRAY)
-    
-    # Initialize face detector
-    face_detector = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
-    
-    # Detect faces
-    faces = face_detector.detectMultiScale(gray, 1.3, 5)
-    
-    if len(faces) == 0:
-        return {"error": "No faces detected in the image"}
-    
-    # List to store recognized students
-    present_students = set()
-    
-    # Compare each detected face with student face images
-    for (x, y, w, h) in faces:
-        face_img = gray[y:y+h, x:x+w]
-        face_img = cv2.resize(face_img, (100, 100))
+    try:
+        # Convert to grayscale
+        gray = cv2.cvtColor(image_array, cv2.COLOR_BGR2GRAY)
         
-        for name, student_face in students.items():
-            # Simple template matching for demonstration
-            # In a production system, you'd use a proper face recognition algorithm
-            result = cv2.matchTemplate(face_img, student_face, cv2.TM_CCOEFF_NORMED)
-            similarity = np.max(result)
+        # Initialize face detector
+        face_detector = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
+        
+        # Detect faces
+        faces = face_detector.detectMultiScale(gray, 1.3, 5)
+        
+        if len(faces) == 0:
+            return {"error": "No faces detected in the image"}
+        
+        # List to store recognized students
+        present_students = set()
+        
+        # Compare each detected face with student face images
+        for (x, y, w, h) in faces:
+            face_img = gray[y:y+h, x:x+w]
+            face_img = cv2.resize(face_img, (100, 100))
             
-            # Threshold for recognition
-            if similarity > 0.6:
-                present_students.add(name)
-    
-    missing_students = set(students.keys()) - present_students
-    return {"missing_students": list(missing_students)}
+            for name, student_face in students.items():
+                # Simple template matching for demonstration
+                # In a production system, you'd use a proper face recognition algorithm
+                result = cv2.matchTemplate(face_img, student_face, cv2.TM_CCOEFF_NORMED)
+                similarity = np.max(result)
+                
+                # Threshold for recognition
+                if similarity > 0.6:
+                    present_students.add(name)
+        
+        missing_students = set(students.keys()) - present_students
+        return {"present_students": list(present_students), "missing_students": list(missing_students)}
+    except Exception as e:
+        return {"error": f"Error processing image: {str(e)}"}
 
 @app.post("/create_folder/")
 async def create_folder(org_name: str = Form(...), folder_name: str = Form(...), files: List[UploadFile] = File(...)):