Update app.py

app.py

@@ -15,7 +15,7 @@ activity_categories = {
     "Working": ["laptop", "computer", "keyboard", "office chair"],
     "Meal Time": ["fork", "spoon", "plate", "food"],
     "Exercise": ["dumbbell", "bicycle", "yoga mat", "treadmill"],
-    "Outdoors": ["car", "tree", "bicycle", "road"],
+    "Outdoors": ["car", "tree", "bicycle", "road","subway","metro"],
     # Add more categories and objects as needed
 }
 
@@ -63,11 +63,14 @@ def generate_journal_with_images(video_path, frame_interval=30):
             # Make predictions using YOLOv10 on the current frame
             results = model.predict(source=frame_rgb, device=device)
             
-            # Extract detected objects
-            detected_objects = [model.names[int(box.cls)] for box in results[0].boxes]
+            # Filter detected objects based on confidence threshold
+            detected_objects = []
+            for box in results[0].boxes:
+                if box.conf >= confidence_threshold:  # Only include objects with confidence >= 0.8
+                    detected_objects.append(model.names[int(box.cls)])
             
-            # Only process frames where objects are detected
-            if detected_objects:  # If there are detected objects in the frame
+            # Only process frames where objects with confidence >= threshold are detected
+            if detected_objects:  # If there are high-confidence detected objects
                 
                 # Plot bounding boxes and labels on the image
                 annotated_frame = results[0].plot()  # Plot detection results on the frame
@@ -93,7 +96,7 @@ def generate_journal_with_images(video_path, frame_interval=30):
     
     cap.release()
     
-    return journal_entries, image_paths
+    return journal_entries, image_paths 
 
 
 def display_journal_with_images(video):