Update app.py

app.py

@@ -28,10 +28,10 @@ DepthwiseConv2D.from_config = classmethod(patched_dwconv_from_config)
 IMG_SIZE = (224, 224)
 CLASS_NAMES = ['Normal', 'Diabetes', 'Glaucoma', 'Cataract', 'AMD', 'Hypertension', 'Myopia', 'Others']
 
-# --- Load model using TFSMLayer ---
+# --- Load model ---
 @st.cache_resource
 def load_model():
-    model_path = "Model"  # Folder path to TF SavedModel
+    model_path = "Model"
     if not os.path.exists(model_path):
         st.error(f"❌ Model folder '{model_path}' not found!")
         st.stop()
@@ -44,8 +44,8 @@ def load_model():
         st.error(f"❌ Error loading model: {str(e)}")
         st.stop()
 
-# --- Preprocessing with visualization ---
-def preprocess_and_show_steps(img):
+# --- Preprocessing function ---
+def preprocess_image(img):
     h, w = img.shape[:2]
     center = (w // 2, h // 2)
     radius = min(center[0], center[1])
@@ -65,26 +65,11 @@ def preprocess_and_show_steps(img):
     sharp = cv2.addWeighted(clahe_img, 4, blur, -4, 128)
 
     resized = cv2.resize(sharp, IMG_SIZE) / 255.0
+    return resized, [img, circ, clahe_img, resized]
 
-    # Show preprocessing stages
-    fig, axs = plt.subplots(1, 4, figsize=(20, 5))
-    axs[0].imshow(img)
-    axs[0].set_title("Original Image")
-    axs[1].imshow(circ)
-    axs[1].set_title("After Circular Crop")
-    axs[2].imshow(clahe_img)
-    axs[2].set_title("After CLAHE")
-    axs[3].imshow(resized)
-    axs[3].set_title("Sharpen + Resize")
-    for ax in axs:
-        ax.axis("off")
-    st.pyplot(fig)
-    plt.close()
-
-    return resized
-
-# --- LIME Explainer ---
+# --- LIME explainer ---
 explainer = lime_image.LimeImageExplainer()
+
 def predict_fn(images):
     images = np.array(images)
     preds = model.predict(images, verbose=0)
@@ -92,66 +77,82 @@ def predict_fn(images):
         preds = list(preds.values())[0]
     return preds
 
-# --- Explanation texts ---
+# --- Explanation text ---
 explanation_text = {
-    'Normal': "Model predicted Normal based on healthy optic disc and macula.",
-    'Diabetes': "Detected retinal blood vessel changes suggestive of Diabetes.",
-    'Glaucoma': "Detected increased cupping in the optic disc indicating Glaucoma.",
-    'Cataract': "Image blur indicated potential Cataract.",
-    'AMD': "Degeneration signs in macula indicate AMD.",
-    'Hypertension': "Blood vessel narrowing/hemorrhages indicate Hypertension.",
-    'Myopia': "Tilted disc and fundus shape suggest Myopia.",
-    'Others': "Non-specific features detected, marked as Others."
+    'Normal': "Healthy optic disc and macula.",
+    'Diabetes': "Retinal vessel changes suggest Diabetes.",
+    'Glaucoma': "Optic disc cupping detected.",
+    'Cataract': "Blurring suggests Cataract.",
+    'AMD': "Degeneration signs in macula.",
+    'Hypertension': "Hemorrhages suggest Hypertension.",
+    'Myopia': "Fundus tilt suggests Myopia.",
+    'Others': "Non-specific features detected."
 }
 
-# --- Display LIME only ---
-def display_lime_visualization(img, true_label, pred_label, pred_idx):
+# --- Display results ---
+def display_all_results(image_name, orig_img, processed_img, stages, pred_label, confidence, pred_idx):
+    st.header(f"🖼️ Image: `{image_name}`")
+
+    # Show preprocessing
+    st.subheader("🔍 Preprocessing Steps")
+    fig, axs = plt.subplots(1, 4, figsize=(20, 5))
+    titles = ["Original", "Circular Crop", "CLAHE", "Sharpen + Resize"]
+    for i, (img, title) in enumerate(zip(stages, titles)):
+        axs[i].imshow(img)
+        axs[i].set_title(title)
+        axs[i].axis('off')
+    st.pyplot(fig)
+    plt.close()
+
+    st.success(f"✅ Prediction: **{pred_label}** with confidence **{confidence:.2f}%**")
+
+    # LIME
     with st.spinner("🟡 LIME Explanation is Loading..."):
         explanation = explainer.explain_instance(
-            image=img,
+            image=processed_img,
             classifier_fn=predict_fn,
             top_labels=1,
             hide_color=0,
             num_samples=1000
         )
         temp, mask = explanation.get_image_and_mask(label=pred_idx, positive_only=True, num_features=10, hide_rest=False)
-
         fig, axs = plt.subplots(1, 2, figsize=(12, 5))
-        axs[0].imshow(img)
-        axs[0].set_title(f"Original\nTrue: {true_label}")
+        axs[0].imshow(processed_img)
+        axs[0].set_title("Processed Image")
         axs[1].imshow(mark_boundaries(temp, mask))
-        axs[1].set_title(f"LIME Explanation\nPred: {pred_label}")
+        axs[1].set_title("LIME Explanation")
         for ax in axs:
-            ax.axis('off')
-
-        summary = explanation_text.get(pred_label, "Model detected features matching this class.")
-        plt.figtext(0.5, 0.01, summary, wrap=True, ha='center', fontsize=10)
-        plt.tight_layout(rect=[0, 0.03, 1, 1])
+            ax.axis("off")
+        plt.figtext(0.5, 0.01, explanation_text.get(pred_label, ""), ha="center", fontsize=10)
         st.pyplot(fig)
         plt.close()
 
-# --- Streamlit UI ---
-st.set_page_config(page_title="🧠 Retina Disease Classifier with LIME", layout="centered")
+# --- App UI ---
+st.set_page_config(page_title="🧠 Retina Disease Classifier", layout="centered")
 st.title("🧠 Retina Disease Classifier with LIME Explanation")
 
 model = load_model()
 
-uploaded_file = st.file_uploader("Upload a retinal image", type=["jpg", "jpeg", "png"])
-if uploaded_file:
-    file_bytes = np.asarray(bytearray(uploaded_file.read()), dtype=np.uint8)
-    bgr_img = cv2.imdecode(file_bytes, cv2.IMREAD_COLOR)
-    rgb_img = cv2.cvtColor(bgr_img, cv2.COLOR_BGR2RGB)
-
-    processed_img = preprocess_and_show_steps(rgb_img)
-
-    input_tensor = np.expand_dims(processed_img, axis=0)
-    preds = model.predict(input_tensor)
-    if isinstance(preds, dict):
-        preds = list(preds.values())[0]
-    pred_idx = np.argmax(preds)
-    pred_label = CLASS_NAMES[pred_idx]
-    confidence = np.max(preds) * 100
-
-    st.success(f"✅ Prediction: **{pred_label}** with confidence **{confidence:.2f}%**")
-
-    display_lime_visualization(processed_img, "Uploaded Image", pred_label, pred_idx)
+uploaded_files = st.file_uploader(
+    "Upload one or more retinal images",
+    type=["jpg", "jpeg", "png"],
+    accept_multiple_files=True
+)
+
+if uploaded_files:
+    for uploaded_file in uploaded_files:
+        file_bytes = np.asarray(bytearray(uploaded_file.read()), dtype=np.uint8)
+        bgr_img = cv2.imdecode(file_bytes, cv2.IMREAD_COLOR)
+        rgb_img = cv2.cvtColor(bgr_img, cv2.COLOR_BGR2RGB)
+
+        processed_img, stages = preprocess_image(rgb_img)
+        input_tensor = np.expand_dims(processed_img, axis=0)
+
+        preds = model.predict(input_tensor)
+        if isinstance(preds, dict):
+            preds = list(preds.values())[0]
+        pred_idx = np.argmax(preds)
+        pred_label = CLASS_NAMES[pred_idx]
+        confidence = np.max(preds) * 100
+
+        display_all_results(uploaded_file.name, rgb_img, processed_img, stages, pred_label, confidence, pred_idx)