Update app.py

app.py

@@ -5,47 +5,42 @@ from ultralytics import YOLO
 
 st.set_page_config(page_title="Suspicious Activity Detection", layout="centered")
 
-# Load YOLOv11 model
+# Load the YOLOv11 model
 @st.cache_resource
 def load_model():
-    return YOLO("yolo11l.pt")
+    return YOLO("yolo11l.pt")  # Ensure your model file is present in the root directory
 
 model = load_model()
 
-# Improved Action Classification Logic
-def classify_action(detections, boxes):
+# ------------------ Intelligent Action Classification Logic ------------------
+def classify_action(detections):
+    """
+    Classifies the activity based on detected object types and confidence scores.
+    Adjusted to better separate 'Sneaking' from 'Peaking'.
+    """
     action_scores = {'Stealing': 0, 'Sneaking': 0, 'Peaking': 0, 'Normal': 0}
     objects = [d[0] for d in detections]
     confidences = [d[1] for d in detections]
 
     has_person = 'person' in objects
     has_bag = any(obj in objects for obj in ['backpack', 'handbag'])
-    low_conf = max(confidences) < 0.5 if confidences else True
+    low_conf = max(confidences) < 0.6 if confidences else True
     few_objects = len(set(objects)) <= 2
-    mostly_person = objects.count('person') > len(objects) * 0.6
-
-    # Analyze bounding box heights (sneaking = crouched or hidden)
-    person_heights = [
-        y2 - y1 for (cls, conf), (x1, y1, x2, y2) in zip(detections, boxes)
-        if cls == 'person'
-    ]
-    avg_height = np.mean(person_heights) if person_heights else 0
-    sneaking_posture = avg_height < 0.4  # Normalized threshold (assuming 0–1 scale)
+    mostly_person = objects.count('person') >= len(objects) * 0.6
 
     if has_person:
         if has_bag:
-            action_scores['Stealing'] += 0.6
-        if 'refrigerator' in objects or 'microwave' in objects:
-            action_scores['Stealing'] += 0.4
-        if low_conf and sneaking_posture:
+            action_scores['Stealing'] += 0.7
+        if low_conf and few_objects:
             action_scores['Sneaking'] += 0.9
-        elif few_objects and mostly_person:
-            action_scores['Peaking'] += 0.7
+        elif mostly_person and few_objects:
+            action_scores['Peaking'] += 0.6
         else:
             action_scores['Normal'] += 0.8
     else:
         action_scores['Normal'] += 1.0
 
+    # Normalize scores
     total = sum(action_scores.values())
     if total > 0:
         for k in action_scores:
@@ -53,25 +48,26 @@ def classify_action(detections, boxes):
 
     return action_scores
 
-# Detection Function
+# ------------------ Detection Function ------------------
 def detect_action(image_path):
     results = model.predict(source=image_path, conf=0.4, iou=0.5, save=False, verbose=False)
     result = results[0]
+
     detections = [
         (model.names[int(cls)], float(conf))
         for cls, conf in zip(result.boxes.cls, result.boxes.conf)
     ]
-    boxes = result.boxes.xyxy.cpu().numpy() / result.orig_shape[0]  # Normalize height
 
     annotated_image = result.plot()
-    action_scores = classify_action(detections, boxes)
+    action_scores = classify_action(detections)
+
     return annotated_image, action_scores
 
-# Streamlit UI
+# ------------------ Streamlit UI ------------------
 st.title("🛡️ Suspicious Activity Detection")
-st.markdown("Upload an image to detect: **Stealing**, **Sneaking**, **Peaking**, or **Normal** behavior.")
+st.markdown("Upload an image to detect if someone is **Stealing**, **Sneaking**, **Peaking**, or acting **Normal**.")
 
-uploaded_file = st.file_uploader("📤 Upload Image", type=["jpg", "jpeg", "png"])
+uploaded_file = st.file_uploader("📸 Upload an image", type=["jpg", "jpeg", "png"])
 
 if uploaded_file:
     image = Image.open(uploaded_file).convert("RGB")
@@ -80,13 +76,14 @@ if uploaded_file:
     temp_path = "/tmp/uploaded.jpg"
     image.save(temp_path)
 
-    with st.spinner("🔍 Analyzing..."):
+    with st.spinner("🔍 Detecting suspicious activity..."):
         detected_image, action_scores = detect_action(temp_path)
 
-    st.image(detected_image, caption="Detection Result", use_column_width=True)
-    st.subheader("📊 Action Scores")
+    st.image(detected_image, caption="🔍 Detection Results", use_column_width=True)
+
+    st.subheader("📊 Action Confidence Scores")
     for action, score in action_scores.items():
         st.write(f"**{action}**: {score:.2%}")
-    
-    final_action = max(action_scores.items(), key=lambda x: x[1])
-    st.success(f"🎯 **Predicted Action:** {final_action[0]} ({final_action[1]:.2%})")
+
+    top_action = max(action_scores.items(), key=lambda x: x[1])
+    st.success(f"🎯 **Predicted Action:** {top_action[0]} ({top_action[1]:.2%} confidence)")