Update cal.py

cal.py

@@ -132,12 +132,38 @@ def predict_image(image_path, model, conf_threshold=0.03):
 
 # Function to calculate detected items and their calories
 def calculate_calories(detection_details):
-    detected_items = []
+    """
+    Calculate calories for detected items, keeping only the highest confidence detection for each unique food item.
     
+    Args:
+        detection_details: List of dictionaries containing detection information
+            Each dict has keys: "class" (food name), "top_confidence" (detection confidence), "bbox"
+            
+    Returns:
+        List of tuples: (food_item, calories, confidence) for unique items with highest confidence
+    """
+    # Dictionary to keep track of highest confidence detection for each food item
+    unique_items = {}
+    
+    # Process each detection
     for det in detection_details:
         item = det["class"]
-        calories = Config.CALORIES_DICT[item]
         confidence = det["top_confidence"]
-        detected_items.append((item, calories, confidence))
+        
+        # Only update if this is the first instance or has higher confidence
+        if item not in unique_items or confidence > unique_items[item]["confidence"]:
+            unique_items[item] = {
+                "calories": Config.CALORIES_DICT[item],
+                "confidence": confidence
+            }
+    
+    # Convert to list of tuples format
+    detected_items = [
+        (item, data["calories"], data["confidence"])
+        for item, data in unique_items.items()
+    ]
+    
+    # Sort by confidence (optional)
+    detected_items.sort(key=lambda x: x[2], reverse=True)
     
     return detected_items