Update main.py

main.py

@@ -17,66 +17,23 @@ app.add_middleware(
 )
 
 # ==========================================
-# 🧠 LOAD MODELS (Classification Only!)
+# LOAD MODELS (Classification Only!)
 # ==========================================
 print("⏳ Loading Models...")
 
 try:
     # 1. Disease Classification Model (Your ultra-smart 96.8% brain)
     disease_model = YOLO('best.pt')
-    print("✅ Goyam Disease Classifier loaded!")
+    print(" Goyam Disease Classifier loaded!")
 except Exception as e:
-    print(f"❌ Error loading Disease Model: {e}")
-
-try:
-    # 2. Plant Gatekeeper (ImageNet 1000-class classifier)
-    gatekeeper_model = YOLO('yolov8n-cls.pt') 
-    print("✅ Plant Gatekeeper loaded!")
-except Exception as e:
-    print(f"❌ Error loading Gatekeeper: {e}")
+    print(f" Error loading Disease Model: {e}")
 
+# Note: The ImageNet Gatekeeper was removed due to macro-photography hallucinations.
+# We now use a Confidence Threshold architecture.
 
 # ==========================================
-# 🛡️ THE BULLETPROOF GATEKEEPER (Allowlist)
+#  RECOMMENDATION MAPPING
 # ==========================================
-def is_likely_plant(image_path):
-    """
-    STRICT ALLOWLIST LOGIC:
-    We check the top 5 things the AI thinks it sees. 
-    If NONE of them are related to nature, agriculture, or plants, we reject the photo.
-    """
-    try:
-        results = gatekeeper_model(image_path, verbose=False)
-        
-        # Get the top 5 predicted classes
-        top5_indices = results[0].probs.top5
-        top5_names = [results[0].names[i].lower() for i in top5_indices]
-        
-        print(f"🧐 Gatekeeper sees: {top5_names}")
-
-        # ✅ ALLOW LIST: Botanical, agricultural, and nature terms
-        allowed_keywords = [
-            'plant', 'leaf', 'grass', 'flower', 'tree', 'fern', 'moss', 'weed', 
-            'crop', 'agriculture', 'field', 'greenhouse', 'pot', 'earth', 'soil', 
-            'vegetation', 'forest', 'valley', 'daisy', 'corn', 'acorn', 'paddy'
-        ]
-
-        # Check if ANY of the top 5 predictions contain our allowed keywords
-        for predicted_item in top5_names:
-            for good_word in allowed_keywords:
-                if good_word in predicted_item:
-                    print(f"✅ Passed: Gatekeeper verified plant matter ('{predicted_item}')")
-                    return True
-
-        # If the loop finishes and didn't find a single nature word, reject it!
-        print(f"🚫 Blocked: No agricultural or plant features detected.")
-        return False
-
-    except Exception as e:
-        print(f"⚠️ Gatekeeper Error: {e}")
-        return True # Fail-safe: let it pass if the gatekeeper crashes
-
-
 def get_recommendation(disease_name):
     recommendations = {
         "Leaf Blast": "Use Tricyclazole 75 WP. Avoid applying excess nitrogen fertilizer.",
@@ -90,10 +47,9 @@ def get_recommendation(disease_name):
             return value
     return "Consult your local agricultural extension officer for treatment."
 
-
 @app.get("/")
 def home():
-    return {"message": "Goyam AI is Running! 🚀"}
+    return {"message": "Goyam AI is Running! "}
 
 @app.post("/predict")
 async def predict(
@@ -108,18 +64,7 @@ async def predict(
         with open(temp_filename, "wb") as buffer:
             shutil.copyfileobj(file.file, buffer)
 
-        # 🛑 STEP 1: RUN THE STRICT GATEKEEPER
-        if not is_likely_plant(temp_filename):
-            return {
-                "filename": file.filename,
-                "disease": "Invalid Image",
-                "confidence": "0%",
-                "recommendation": "This image does not appear to be a plant or paddy field. Please upload a clear photo of a rice leaf.",
-                "latitude": float(latitude) if latitude else None,
-                "longitude": float(longitude) if longitude else None
-            }
-
-        # ✅ STEP 2: RUN DISEASE CLASSIFICATION
+        #  STEP 1: RUN THE EXPERT DISEASE MODEL
         results = disease_model(temp_filename, verbose=False)
         
         # Extract the highest probability prediction
@@ -127,6 +72,21 @@ async def predict(
         confidence_score = float(results[0].probs.top1conf)
         detected_name = results[0].names[top_idx]
         
+        #  STEP 2: THE CONFIDENCE THRESHOLD GATEKEEPER
+        # If your model is less than 50% sure, it's probably looking at a car, a dog, or a random object.
+        if confidence_score < 0.50:
+            print(f" Blocked: Confidence too low ({confidence_score*100:.1f}%). Likely not a clear paddy leaf.")
+            return {
+                "filename": file.filename,
+                "disease": "Invalid Image",
+                "confidence": f"{int(confidence_score * 100)}%",
+                "recommendation": "This image does not appear to be a clear photo of a paddy field. Please upload a focused photo of a rice leaf.",
+                "latitude": float(latitude) if latitude else None,
+                "longitude": float(longitude) if longitude else None
+            }
+
+        # STEP 3: SUCCESSFUL DIAGNOSIS
+        print(f" Passed: Diagnosed {detected_name} with {confidence_score*100:.1f}% confidence.")
         response_data = {
             "filename": file.filename,
             "disease": detected_name,
@@ -139,7 +99,7 @@ async def predict(
         return response_data
 
     except Exception as e:
-        print(f"❌ API Error: {e}")
+        print(f" API Error: {e}")
         return {"error": str(e)}
         
     finally: