Update app.py

app.py

@@ -7,7 +7,7 @@ import zipfile
 import shutil
 import json
 from datetime import datetime
-from predict_task1 import Predictor
+from predict_task2 import Predictor
 from id_mapping import mapping
 from show_stitched import *
 import cv2
@@ -48,7 +48,7 @@ os.makedirs(IMAGES_FOLDER, exist_ok=True)
 # Initialize class mapping file
 if not os.path.exists(CLASS_MAPPING_FILE):
     # Create initial class mapping from id_mapping.py
-    reverse_mapping = {str(v): k for k, v in mapping.items() if v != -1 and k is not None}
+    reverse_mapping = {str(v): k for k, v in mapping.items() if v not in [-999] and k is not None}
     with open(CLASS_MAPPING_FILE, 'w', encoding='utf-8') as f:
         json.dump(reverse_mapping, f, indent=2, ensure_ascii=False)
 
@@ -70,7 +70,7 @@ def get_compatible_class_id(yolo_class_name):
         "circle": 40, "Bullseye": -1, "bullseye": -1
     }
     
-    return yolo_to_id_mapping.get(yolo_class_name, mapping.get(yolo_class_name, -1))
+    return yolo_to_id_mapping.get(yolo_class_name, mapping.get(yolo_class_name, -999))
 
 def load_class_mapping():
     """Load class mapping from JSON file"""
@@ -290,6 +290,50 @@ def process_file(file_path, direction, task_type, filename):
     
     # Perform prediction
     class_name, results, detection_id = predictor.predict_id(file_path, task_type)
+    
+    # For TASK_2, apply priority-based selection AFTER getting all detections
+    if task_type == "TASK_2" and results and results[0].boxes is not None and len(results[0].boxes) > 0:
+        priority_classes = ['right', 'left', 'up', 'down']
+        detections_list = []
+        
+        boxes = results[0].boxes
+        for i in range(len(boxes)):
+            detected_class = results[0].names[int(boxes.cls[i])]
+            confidence = float(boxes.conf[i])
+            yolo_class_id = int(boxes.cls[i])
+            
+            print(f"[APP.PY] Detection {i}: {detected_class} (confidence: {confidence:.2f}, class_id: {yolo_class_id})")
+            
+            # Assign priority based on class name
+            if detected_class in priority_classes:
+                priority = len(priority_classes) - priority_classes.index(detected_class)
+            elif detected_class == 'Bullseye' or detected_class == 'bullseye':
+                priority = -10  # Lowest priority for bullseye
+            else:
+                priority = -1  # Lower priority for other symbols
+            
+            detections_list.append({
+                'index': i,
+                'class_name': detected_class,
+                'confidence': confidence,
+                'priority': priority,
+                'yolo_class_id': yolo_class_id
+            })
+        
+        if detections_list:
+            # Sort by priority (descending), then by confidence (descending)
+            detections_list.sort(key=lambda x: (x['priority'], x['confidence']), reverse=True)
+            
+            print(f"\n[APP.PY] Sorted detections:")
+            for det in detections_list:
+                print(f"  - {det['class_name']}: priority={det['priority']}, confidence={det['confidence']:.2f}")
+            
+            # Override with highest priority detection
+            selected = detections_list[0]
+            class_name = selected['class_name']
+            detection_id = selected['index']
+            print(f"\n[APP.PY] ✓ Final selection: {class_name} (priority: {selected['priority']}, confidence: {selected['confidence']:.2f})")
+    
     # Use compatible mapping function
     class_id = str(get_compatible_class_id(class_name))
     
@@ -591,7 +635,7 @@ HTML_TEMPLATE = """
                         <label for="task_type">Task Type:</label>
                         <select id="task_type" name="task_type" required>
                             <option value="TASK_1">Task 1</option>
-                            <option value="TASK_2">Task 2</option>
+                            <option value="TASK_2" selected>Task 2</option>
                         </select>
                     </div>
                     <div class="form-group">
@@ -763,7 +807,7 @@ HTML_TEMPLATE = """
                     `;
                     
                     if (numObstacles && numObstacles !== '0') {
-                        if (predictedId === '-1') {
+                        if (predictedId === '-999') {
                             successMessage += `<br><span style="color: #ffc107;">⚠️ Obstacle ${numObstacles} not saved - No valid detection</span>`;
                         } else {
                             successMessage += `<br><span style="color: #28a745;">✓ Saved as Obstacle ${numObstacles}</span>`;
@@ -1104,7 +1148,7 @@ OBSTACLES_HTML_TEMPLATE = """
             if (existingCount === 0) {
                 statusEl.style.backgroundColor = '#fff3cd';
                 statusEl.style.color = '#856404';
-                statusEl.textContent = 'No obstacles detected yet. Upload images with NUM_OBSTACLES parameter (1-8) and valid detections (ID ≠ -1).';
+                statusEl.textContent = 'No obstacles detected yet. Upload images with NUM_OBSTACLES parameter (1-8) and valid detections (ID ≠ -999).';
             } else {
                 statusEl.style.backgroundColor = '#d4edda';
                 statusEl.style.color = '#155724';
@@ -1164,7 +1208,7 @@ def image_predict():
     
     # Get parameters from both old and new format
     direction = request.form.get('direction', 'north')
-    task_type = request.form.get('task_type', 'TASK_1')
+    task_type = request.form.get('task_type', 'TASK_2')
     num_obstacles = request.form.get('NUM_OBSTACLES', '0')  # Support NUM_OBSTACLES parameter
     
     # Try to parse filename format: "<timestamp>_<obstacle_id>_<signal>.jpeg"
@@ -1200,14 +1244,14 @@ def image_predict():
     class_id, detection_result = process_file(file_path, signal, task_type, filename)
     
     # Handle NUM_OBSTACLES parameter for obstacle display
-    # Only save if detection is valid (image_id != '-1')
+    # Only save if detection is valid (image_id != '-999')
     if (num_obstacles and num_obstacles.isdigit() and 1 <= int(num_obstacles) <= 8 
-        and detection_result and detection_result.get('image_id') != '-1'):
+        and detection_result and detection_result.get('image_id') != '-999'):
         save_obstacle_image(detection_result, int(num_obstacles))
         print(f"Obstacle {num_obstacles} saved with valid detection (ID: {detection_result.get('image_id')})")
     elif (num_obstacles and num_obstacles.isdigit() and 1 <= int(num_obstacles) <= 8 
-          and detection_result and detection_result.get('image_id') == '-1'):
-        print(f"Obstacle {num_obstacles} NOT saved - invalid detection (ID: -1)")
+          and detection_result and detection_result.get('image_id') == '-999'):
+        print(f"Obstacle {num_obstacles} NOT saved - invalid detection (ID: -999)")
     elif num_obstacles and num_obstacles.isdigit() and 1 <= int(num_obstacles) <= 8:
         print(f"Obstacle {num_obstacles} NOT saved - no detection result")