update space

.env

@@ -3,3 +3,4 @@
 MONGO_URI="your_mongo_db_uri"
 EMAIL_USER="your_email_username"
 EMAIL_PASS="your_email_password"
+TF_ENABLE_ONEDNN_OPTS=0

.idea/Generative_AI_with_poultry_disease_detection_system_v2.iml

@@ -8,7 +8,4 @@
     <orderEntry type="sourceFolder" forTests="false" />
     <orderEntry type="library" name="jquery-3.6.0" level="application" />
   </component>
-  <component name="PackageRequirementsSettings">
-    <option name="removeUnused" value="true" />
-  </component>
 </module>

app.py

@@ -5,21 +5,26 @@ from fastapi.templating import Jinja2Templates
 from fastapi.responses import HTMLResponse
 from fastapi.staticfiles import StaticFiles
 from routes.authentication import auth_router
+from routes.disease_detection import disease_router  # Import the new router
 import os
+import tensorflow as tf
+
+
+print("Num GPUs Available: ", len(tf.config.list_physical_devices('GPU')))
 
-# Initialize FastAPI app and template directory
 app = FastAPI()
 templates = Jinja2Templates(directory="templates")
 
-# Mount static files if the directory exists
+# Mount static files if directory exists
 static_dir = "static"
 if os.path.isdir(static_dir):
     app.mount("/static", StaticFiles(directory=static_dir), name="static")
 else:
     raise HTTPException(status_code=500, detail="Static directory not found.")
 
-# Include authentication routes
+# Include routers
 app.include_router(auth_router, prefix="/auth", tags=["Authentication"])
+app.include_router(disease_router, prefix="/disease", tags=["Disease Detection"])  # New route
 
 @app.get("/", response_class=HTMLResponse)
 async def landing_page(request: Request):

requirements.txt

@@ -1,10 +1,13 @@
-fastapi~=0.115.4
-passlib[bcrypt]~=1.7.4
-pydantic[email]~=2.9.2
-pymongo~=4.9.2
-python-dotenv~=1.0.1
-python-jose[cryptography]~=3.3.0
-starlette~=0.41.2
+fastapi
+passlib[bcrypt]
+pydantic[email]
+pymongo
+python-dotenv
+python-jose[cryptography]
+starlette
 uvicorn
 jinja2
-python-multipart
+python-multipart
+numpy
+pillow
+tensorflow[and-cuda]~=2.9

routes/disease_detection.py

@@ -0,0 +1,46 @@
+# routes/disease_detection.py
+
+from fastapi import APIRouter, UploadFile, File, HTTPException
+from fastapi.responses import JSONResponse
+from PIL import Image
+import io
+import numpy as np
+from services.image_preprocessing import preprocess_image
+from services.disease_detection_service import load_disease_model
+
+disease_router = APIRouter()
+model = load_disease_model()
+
+
+@disease_router.post("/detect_disease", response_class=JSONResponse)
+async def detect_disease(file: UploadFile = File(...)):
+    """
+    Detect disease from uploaded poultry fecal image.
+
+    Parameters:
+        file (UploadFile): Uploaded image file to analyze.
+
+    Returns:
+        JSONResponse: Prediction result indicating disease probability or classification.
+    """
+    if file.content_type not in ["image/jpeg", "image/png"]:
+        raise HTTPException(status_code=400, detail="Invalid file type. Please upload a JPEG or PNG image.")
+
+    # Load image and preprocess
+    image_data = await file.read()
+    image = Image.open(io.BytesIO(image_data))
+    processed_image = preprocess_image(image)
+
+    # Predict using the loaded model
+    prediction = model.predict(processed_image)
+    predicted_class = np.argmax(prediction, axis=1).item()
+    confidence = prediction[0][predicted_class] * 100
+
+    # Interpret results (assume labels for example: 0='Healthy', 1='Infected')
+    label_map = {0: "Healthy", 1: "Infected"}
+    result = {
+        "prediction": label_map.get(predicted_class, "Unknown"),
+        "confidence": f"{confidence:.2f}%"
+    }
+
+    return JSONResponse(content=result)

services/convert_model.py

@@ -0,0 +1,12 @@
+# convert_model.py
+
+import tensorflow as tf
+
+MODEL_PATH = "models/Final_Chicken_disease_model.h5"
+NEW_MODEL_PATH = "models/Final_Chicken_disease_model"
+
+# Load the .h5 model
+model = tf.keras.models.load_model(MODEL_PATH)
+
+# Save it in the SavedModel format
+model.save(NEW_MODEL_PATH)

services/disease_detection_service.py

@@ -0,0 +1,35 @@
+# services/disease_detection_service.py
+
+import tensorflow as tf
+from tensorflow.keras.models import load_model
+import numpy as np
+from PIL import Image
+import os
+
+# Enable memory growth for GPUs
+gpus = tf.config.list_physical_devices('GPU')
+if gpus:
+    try:
+        for gpu in gpus:
+            tf.config.experimental.set_memory_growth(gpu, True)
+        print("Memory growth enabled for GPU.")
+    except RuntimeError as e:
+        print(f"Error setting memory growth: {e}")
+
+
+MODEL_PATH = "models/Final_Chicken_disease_model.h5"
+model = None
+
+def load_disease_model():
+    """Load the trained disease detection model with GPU support if available."""
+    global model
+    if model is None:
+        # Use GPU if available
+        if tf.config.list_physical_devices('GPU'):
+            print("GPU is available. Using GPU for model inference.")
+            with tf.device('/GPU:0'):
+                model = load_model(MODEL_PATH, custom_objects={'BatchNormalization': tf.keras.layers.BatchNormalization})
+        else:
+            print("GPU not available. Using CPU for model inference.")
+            model = load_model(MODEL_PATH, custom_objects={'BatchNormalization': tf.keras.layers.BatchNormalization})
+    return model

services/image_preprocessing.py

@@ -0,0 +1,22 @@
+# services/image_preprocessing.py
+
+from PIL import Image
+import numpy as np
+
+
+def preprocess_image(image: Image.Image, target_size=(224, 224)):
+    """
+    Preprocesses the uploaded image for disease detection.
+
+    Parameters:
+        image (PIL.Image.Image): Input image to preprocess.
+        target_size (tuple): Target size for resizing the image.
+
+    Returns:
+        np.ndarray: Preprocessed image ready for model input.
+    """
+    image = image.convert("RGB")  # Ensure 3 color channels
+    image = image.resize(target_size)  # Resize image
+    image_array = np.array(image) / 255.0  # Normalize to [0, 1]
+    image_array = np.expand_dims(image_array, axis=0)  # Add batch dimension
+    return image_array