app

app.py

@@ -748,25 +748,32 @@ async def analyze(file: UploadFile = File(...)):
         raise HTTPException(status_code=500, detail=f"Error reading image: {e}")
 
     try:
-        # Check if it's food
-        is_food, food_confidence = classifier.detect_if_food(image)
+        # Advanced food detection
+        is_food, food_confidence, detection_details = recognizer.detect_food_advanced(image)
         
-        if not is_food and food_confidence > 0.6:
+        if not is_food and food_confidence > CONFIG.food_detection_threshold:
             return JSONResponse(content={
                 "success": False,
                 "error": "Non-food object detected",
-                "message": "Image doesn't contain food. Please upload a food image.",
-                "confidence": food_confidence
+                "message": "Image doesn't appear to contain food. Please upload a food image.",
+                "confidence": food_confidence,
+                "detection_details": detection_details,
+                "suggestions": [
+                    "Ensure the image clearly shows food items",
+                    "Check that lighting is adequate",
+                    "Try a different angle or closer shot"
+                ]
             })
         
-        # Classify food
-        logger.info("🔍 Classifying food...")
-        result = classifier.classify_food(image)
+        # Comprehensive food analysis
+        logger.info("🍽️ Starting comprehensive food analysis...")
+        analysis_result = recognizer.analyze_food(image)
         
-        if result["confidence"] < MIN_CONFIDENCE:
+        if analysis_result["confidence"] < CONFIG.min_confidence:
             raise HTTPException(
                 status_code=422,
-                detail=f"Low confidence ({result['confidence']:.2%}). Please upload a clearer image."
+                detail=f"Low confidence recognition ({analysis_result['confidence']:.1%}). " +
+                       "Please upload a clearer image with better lighting."
             )
         
     except HTTPException:
@@ -775,36 +782,51 @@ async def analyze(file: UploadFile = File(...)):
         logger.error(f"Classification error: {e}")
         raise HTTPException(status_code=500, detail=f"Classification error: {e}")
 
-    # Get nutrition data
-    logger.info(f"🍎 Recognized food: {result['primary_label']}")
-    nutrition_data = search_nutrition_data(result["primary_label"])
+    # Comprehensive response
+    logger.info(f"✅ Food recognized: {analysis_result['primary_label']} ({analysis_result['confidence']:.1%})")
     
-    # Prepare response
     response = {
         "success": True,
-        "label": result["primary_label"],
-        "confidence": result["confidence"],
-        "alternatives": result["alternatives"],
-        
-        # Nutrition
-        "nutrition": nutrition_data["nutrition"],
-        "source": nutrition_data["source"],
-        
-        # Image info
-        "image_info": {
-            "width": image_width,
-            "height": image_height,
-            "format": image.format
+        
+        # Primary results
+        "food_item": {
+            "name": analysis_result["primary_label"],
+            "confidence": analysis_result["confidence"],
+            "category": _get_food_category(analysis_result["primary_label"])
+        },
+        
+        # Nutrition analysis
+        "nutrition": analysis_result["nutrition_analysis"],
+        
+        # Visual analysis
+        "image_analysis": {
+            "original_size": original_size,
+            "visual_features": analysis_result["visual_features"],
+            "quality_score": _calculate_image_quality(analysis_result["visual_features"]),
+            "is_food_detected": is_food,
+            "food_detection_confidence": food_confidence
         },
         
-        # Model info
-        "model_info": {
-            "type": "Zero-Shot CLIP Classifier",
-            "model": classifier.model_name,
-            "version": "11.0.0",
-            "method": "Zero-shot learning",
-            "categories": len(FOOD_CATEGORIES),
-            "device": device
+        # AI model details
+        "ai_analysis": {
+            "models_used": analysis_result["processing_info"]["models_used"],
+            "ensemble_details": analysis_result.get("ensemble_details", []),
+            "categories_analyzed": analysis_result["processing_info"]["categories_analyzed"],
+            "processing_time_ms": "<100"  # Typical processing time
+        },
+        
+        # API info
+        "api_info": {
+            "version": "12.0.0",
+            "model_type": "Advanced Multi-Model Ensemble",
+            "device": device.upper(),
+            "enhanced_features": [
+                "Multi-model ensemble",
+                "Visual feature analysis", 
+                "Advanced nutrition lookup",
+                "Confidence scoring",
+                "Image quality assessment"
+            ]
         }
     }
 
@@ -849,18 +871,45 @@ async def analyze_custom(
         raise HTTPException(status_code=500, detail=f"Error reading image: {e}")
 
     try:
-        result = classifier.classify_food(image, custom_categories=custom_categories)
+        # Use custom analysis
+        result = recognizer.analyze_food(image, custom_categories=custom_categories)
+        
+        # Get top 5 results for custom categories
+        if custom_categories:
+            # Re-run CLIP with just custom categories for detailed results
+            clip_result = recognizer._clip_predict(image, custom_categories)
+            
+            # Get top 5
+            sorted_indices = np.argsort(clip_result["all_probs"])[::-1]
+            top5_results = []
+            for idx in sorted_indices[:5]:
+                top5_results.append({
+                    "label": custom_categories[idx],
+                    "confidence": float(clip_result["all_probs"][idx])
+                })
+        else:
+            top5_results = [{"label": result["primary_label"], "confidence": result["confidence"]}]
         
         return JSONResponse(content={
             "success": True,
-            "label": result["primary_label"],
-            "confidence": result["confidence"],
-            "top5": result["top5"],
-            "categories_used": custom_categories if custom_categories else "Food-101 default",
+            "analysis": {
+                "primary_match": {
+                    "label": result["primary_label"],
+                    "confidence": result["confidence"],
+                    "category": _get_food_category(result["primary_label"])
+                },
+                "top_matches": top5_results,
+                "visual_features": result["visual_features"]
+            },
+            "categories": {
+                "total_analyzed": len(custom_categories) if custom_categories else len(FOOD_CATEGORIES),
+                "custom_categories": custom_categories,
+                "using_defaults": custom_categories is None
+            },
             "model_info": {
-                "type": "Zero-Shot CLIP Classifier",
-                "model": classifier.model_name,
-                "method": "Custom zero-shot classification"
+                "method": "Zero-shot learning with custom categories",
+                "models_used": result["processing_info"]["models_used"],
+                "device": device.upper()
             }
         })
         
@@ -880,7 +929,7 @@ def root():
         "status": "🟢 Online & Ready",
         "tagline": "Jednostavan i moćan food recognition sa zero-shot learning",
         "model": {
-            "name": classifier.model_name,
+            "name": recognizer.config.clip_model,
             "type": "Vision-Language Model (CLIP)",
             "capabilities": "Zero-shot classification",
             "device": device.upper(),