Update app.py

app.py

@@ -9,7 +9,7 @@ from googleapiclient.http import MediaIoBaseDownload, MediaFileUpload
 from PIL import Image
 import torch
 import io
-import face_recognition  # Use advanced face recognition model for better accuracy
+from deepface import DeepFace  # Use DeepFace for advanced face recognition
 
 # Set up logging
 logging.basicConfig(level=logging.DEBUG)
@@ -42,13 +42,13 @@ cphotos_folder_id = '1DGeRqRbCPcfLDdEgP0h5fyX-MF8EQ8AH'
 suspects_folder_id = '1N3RMhVD0OygeufLPYod6IYLtqzvlm3Jv'
 
 # Limit the number of threads for scalability
-MAX_THREADS = 10
+MAX_THREADS = 5
 
 def list_files_in_folder(folder_id):
     logging.debug(f"Listing files in folder ID: {folder_id}")
     query = f"'{folder_id}' in parents and trashed=false"
     try:
-        results = drive_service.files().list(q=query, pageSize=100, fields="files(id, name, mimeType)").execute()
+        results = drive_service.files().list(q=query, pageSize=50, fields="files(id, name, mimeType)").execute()
         files = results.get('files', [])
         if not files:
             logging.warning(f"No files found in folder ID: {folder_id}. Ensure that the folder has files and the service account has access.")
@@ -79,23 +79,19 @@ def download_file(file_id, file_name):
         logging.error(f"Error downloading file ID: {file_id} - {e}")
         return None
 
-def extract_face_encoding(image_path):
+def extract_face_embedding(image_path):
     try:
-        image = face_recognition.load_image_file(image_path)
-        face_encodings = face_recognition.face_encodings(image)
-        if face_encodings:
-            return face_encodings[0]
-        else:
-            logging.warning(f"No face detected in image: {image_path}")
-            return None
+        # Use DeepFace to extract embeddings
+        embedding = DeepFace.represent(img_path=image_path, model_name='VGG-Face')[0]["embedding"]
+        return embedding
     except Exception as e:
         logging.error(f"Error extracting face encoding for {image_path}. Error: {e}")
         return None
 
 def calculate_similarity(encoding1, encoding2):
-    distance = face_recognition.face_distance([encoding1], encoding2)[0]
-    similarity = 1 - distance  # Convert distance to similarity
-    return similarity
+    # Use cosine similarity for comparison
+    similarity = DeepFace.find(img_path1=encoding1, img_path2=encoding2, distance_metric='cosine')[0]
+    return 1 - similarity  # Convert distance to similarity
 
 def compare_faces_with_encodings(encoding_cctv, aadhar_encodings, threshold=0.7):
     for aadhar_name, aadhar_encoding in aadhar_encodings:
@@ -110,6 +106,7 @@ def batch_process_images(aadhar_folder_id, cphotos_folder_id, suspects_folder_id
     aadhar_files = list_files_in_folder(aadhar_folder_id)
     cphotos_files = list_files_in_folder(cphotos_folder_id)
 
+    # Extract Aadhar encodings concurrently
     aadhar_encodings = []
     with ThreadPoolExecutor(max_workers=MAX_THREADS) as executor:
         future_to_aadhar = {executor.submit(download_file, file['id'], file['name']): file for file in aadhar_files}
@@ -118,7 +115,7 @@ def batch_process_images(aadhar_folder_id, cphotos_folder_id, suspects_folder_id
             try:
                 file_path = future.result()
                 if file_path:
-                    encoding = extract_face_encoding(file_path)
+                    encoding = extract_face_embedding(file_path)
                     if encoding is not None:
                         aadhar_encodings.append((file['name'], encoding))
             except Exception as e:
@@ -128,6 +125,7 @@ def batch_process_images(aadhar_folder_id, cphotos_folder_id, suspects_folder_id
         logging.warning("No valid Aadhar face encodings found. Exiting.")
         return
 
+    # Extract CCTV encodings and compare concurrently
     unmatched_files = []
     with ThreadPoolExecutor(max_workers=MAX_THREADS) as executor:
         future_to_cphoto = {executor.submit(download_file, file['id'], file['name']): file for file in cphotos_files}
@@ -138,7 +136,7 @@ def batch_process_images(aadhar_folder_id, cphotos_folder_id, suspects_folder_id
                 if not file_path:
                     continue
 
-                encoding_cctv = extract_face_encoding(file_path)
+                encoding_cctv = extract_face_embedding(file_path)
                 if encoding_cctv is None:
                     continue
 
@@ -149,6 +147,7 @@ def batch_process_images(aadhar_folder_id, cphotos_folder_id, suspects_folder_id
             except Exception as e:
                 logging.error(f"Error processing CCTV file {file['name']}: {e}")
 
+    # Upload unmatched files concurrently
     with ThreadPoolExecutor(max_workers=MAX_THREADS) as executor:
         upload_futures = [executor.submit(upload_file, file_path, suspects_folder_id) for file_path in unmatched_files]
         for future in as_completed(upload_futures):