Update app.py

app.py

@@ -2,22 +2,49 @@ import streamlit as st
 from image_classifier import classify_image  # Import from image_classification.py
 
 # Title and description for your app
-st.title("Image Classifier")
-st.write("This app classifies uploaded images using a pre-trained TensorFlow model.")
+st.title("AI vs. Human Art Classifier")
+st.write("This app classifies uploaded images as AI-generated or human-made art.")
 
 # Allow users to upload an image
 uploaded_image = st.file_uploader("Upload an Image", type=["jpg", "jpeg", "png"])
 
 if uploaded_image is not None:
-  # Convert uploaded image to a format suitable for classification
-  image = cv2.imdecode(np.frombuffer(uploaded_image.read(), np.uint8), cv2.IMREAD_COLOR)  # Assuming OpenCV is used for conversion
-
-  # Perform classification using your function
-  try:
+  # Check if data is available in the uploaded image
+  if uploaded_image.read():
+    try:
+      # Attempt OpenCV decoding (assuming OpenCV is installed)
+      image = cv2.imdecode(np.frombuffer(uploaded_image.read(), np.uint8), cv2.IMREAD_COLOR)
       predicted_label, probability = classify_image(image)
       st.write("Classified as:")
       st.write(f"- {predicted_label}: {probability:.4f}")
-  except Exception as e:  # Catch potential errors during classification
-      st.error(f"Error during classification: {e}")
+    except Exception as e:
+      # Handle potential OpenCV errors
+      st.error(f"Error using OpenCV: {e}")
+      # Fallback to Pillow if OpenCV fails
+      try:
+        from PIL import Image
+        image = Image.open(uploaded_image)
+        image = image.convert('RGB')  # Convert to RGB format
+        predicted_label, probability = classify_image(image)  # Pass the image to your function
+        st.write("Classified as (using Pillow):")
+        st.write(f"- {predicted label}: {probability:.4f}")
+      except Exception as e:
+        st.error(f"Error using Pillow for fallback: {e}")
+  else:
+    st.error("Please upload a valid image file.")
 else:
-  st.write("Upload an image to classify it.")
+  st.write("Upload an image to classify it.")
+
+# Additional Information (Optional)
+
+# Based on the classification results, you might want to:
+# - Display a confidence score (probability) for the classification.
+# - Provide a brief explanation of the factors considered for classification.
+# - Offer resources for learning more about AI-generated art.
+
+# Example:
+if predicted_label == "AI-generated art":
+  confidence = probability * 100
+  st.write(f"Confidence level: {confidence:.2f}%")
+  st.write("AI-generated art often exhibits certain stylistic elements or patterns that can be identified by machine learning models. However, human-made art can also incorporate similar elements, and perfect accuracy cannot be guaranteed.")
+  st.write("Learn more about AI-generated art: https://en.wikipedia.org/wiki/Artificial_intelligence_art")