Update app.py

app.py

@@ -4,114 +4,75 @@ import numpy as np
 from PIL import Image
 from mtcnn import MTCNN
 import cv2
-import os
-
-import subprocess
-
-# Define the package you want to install
-package_name = "opencv-python"
-package_name1 = "mtcnn"
-
-# Run the pip install command within your Python script
-subprocess.call(['pip', 'install', package_name])
-subprocess.call(['pip', 'install', package_name1])
+import io
 
 # Load the model (ensure correct path for loading)
+# For Hugging Face Spaces, ensure the model file is included in the repository or loaded from a URL
 model = tf.keras.models.load_model('oily_dry.h5')
 
 # Load the MTCNN face detection model
 mtcnn = MTCNN()
 
-def detect_and_process_skin(image_path, output_folder):
-    """Detects faces in an image, crops the skin region, and saves the processed image."""
-    # Read the input image
-    img = cv2.imread(image_path)
-    img_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
+def detect_and_process_skin(image, output_folder=None):
+    """Detects faces in an image, crops the skin region, and prepares it for classification."""
+    # Convert PIL Image to numpy array
+    img_np = np.array(image)
+    img_rgb = cv2.cvtColor(img_np, cv2.COLOR_BGR2RGB)
 
     # Detect faces in the image
     detections = mtcnn.detect_faces(img_rgb)
 
-    # Check if any faces were detected
     if detections:
         x, y, width, height = detections[0]['box']
 
         # Crop the face region
-        face_img = img[y:y+height, x:x+width]
+        face_img = img_np[y:y+height, x:x+width]
 
-        # Convert face_img to RGB format for processing
+        # Convert cropped face to PIL Image for compatibility with model preprocessing
         pil_img = Image.fromarray(cv2.cvtColor(face_img, cv2.COLOR_BGR2RGB))
 
-        # Save the processed image in the original color format
-        output_path = os.path.join(output_folder, os.path.basename(image_path))
-        cv2.imwrite(output_path, cv2.cvtColor(np.array(pil_img), cv2.COLOR_RGB2BGR))
-
-        return output_path
-
+        return pil_img
     else:
-        return image_path # Return None if no face was detected
-
-def classify_image(image_path):
-    """Loads, preprocess, and classifies the image."""
-    # Process the image using face detection and background removal
-    processed_image_path = detect_and_process_skin(image_path, "processed_images")
+        # Return the original image if no face was detected
+        return image
 
-    if processed_image_path:
-        # Load and resize the processed image
-        image = Image.open(processed_image_path).resize((224, 224))
+def classify_image(image):
+    """Classifies the image after processing."""
+    processed_image = detect_and_process_skin(image)
 
-        # Preprocess the image for the model
-        image = tf.keras.preprocessing.image.img_to_array(image) / 255.0
-        image = np.expand_dims(image, axis=0)
+    # Resize and preprocess the image for the model
+    processed_image = processed_image.resize((224, 224))
+    image_array = tf.keras.preprocessing.image.img_to_array(processed_image) / 255.0
+    image_array = np.expand_dims(image_array, axis=0)
 
-        # Classify the image
-        predictions = model.predict(image)
+    # Classify the image
+    predictions = model.predict(image_array)
+    predicted_class = np.argmax(predictions)
+    percentages = predictions[0] * 100
 
-        # Process predictions for clarity
-        predicted_class = np.argmax(predictions)
-        percentages = predictions[0] * 100
-        dry_percentage, normal_percentage, oily_percentage = percentages
-
-        # Ensure percentages are within 0-100
-        dry_percentage = max(0, min(dry_percentage, 100))
-        oily_percentage = max(0, min(oily_percentage, 100))
-        normal_percentage = max(0, min(normal_percentage, 100))
-
-        return predicted_class, dry_percentage, oily_percentage, normal_percentage
-    else:
-        return None, None, None, None  # Return Nones if no face was detected
+    return predicted_class, *percentages
 
 def app():
-    st.title("Oily/Dry Skin Level Predictor ")
-    st.write("Coded by Manith Jayaba")
+    st.title("Oily/Dry Skin Level Predictor")
 
-    st.write("This app can measure the oiliness and dryness of your skin")
-
-    # Get the image file
     image_file = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])
 
-    # Classify and display the result
     if image_file is not None:
-        image_path = image_file.name
-        with open(image_path, "wb") as f:
-            f.write(image_file.getbuffer())
-
-        predicted_class, dry_percentage, oily_percentage, normal_percentage = classify_image(image_path)
-
-        if predicted_class is not None and os.path.exists(os.path.join("processed_images", image_path)):
-            st.image(os.path.join("processed_images", image_path), width=250)
+        # Convert the file to an image
+        image = Image.open(image_file)
+        predicted_class, dry_percentage, oily_percentage, normal_percentage = classify_image(image)
 
-            # Display progress bars
-            st.progress(int(dry_percentage), text=f"Dry Skin: {dry_percentage:.2f}%")
-            st.progress(int(oily_percentage), text=f"Oily Skin: {oily_percentage:.2f}%")
-            st.progress(int(normal_percentage), text=f"Normal Skin: {normal_percentage:.2f}%")
+        # Convert processed image to bytes for display
+        processed_image_bytes = io.BytesIO()
+        image.save(processed_image_bytes, format='JPEG')
+        processed_image_bytes.seek(0)
 
-        else:
-            st.image(image_path, width=250)
-            st.error("No face detected in the image. Results can be inaccurate")
+        st.image(processed_image_bytes, width=250)
 
-            st.progress(int(dry_percentage), text=f"Dry Skin: {dry_percentage:.2f}%")
-            st.progress(int(oily_percentage), text=f"Oily Skin: {oily_percentage:.2f}%")
-            st.progress(int(normal_percentage), text=f"Normal Skin: {normal_percentage:.2f}%")
+        # Display the classification results
+        st.write(f"Dry Skin: {dry_percentage:.2f}%")
+        st.write(f"Oily Skin: {oily_percentage:.2f}%")
+        st.write(f"Normal Skin: {normal_percentage:.2f}%")
 
 if __name__ == "__main__":
     app()