Update app.py

app.py

@@ -1,6 +1,9 @@
 import streamlit as st
 from transformers import pipeline
 from PIL import Image, ImageDraw
+import torch
+from typing import List, Dict
+import time
 
 st.set_page_config(
     page_title="Fraktur Detektion",
@@ -8,39 +11,90 @@ st.set_page_config(
     initial_sidebar_state="collapsed"
 )
 
+# CSS avec animations
 st.markdown("""
 <style>
     .stApp {
+        background-color: #f8fafc !important;
         padding: 0 !important;
-        height: 100vh !important;
-        overflow: hidden !important;
     }
-
+    
     .block-container {
-        padding: 0.25rem !important;
+        padding: 0.5rem !important;
         max-width: 100% !important;
     }
-
-    .stImage > img {
-        max-height: 150px !important;
-        object-fit: contain !important;
-    }
-
-    h2, h3 {
-        font-size: 0.9rem !important;
-    }
-
-    .result-box {
-        font-size: 0.8rem !important;
-        margin: 0.2rem 0 !important;
-    }
-
-    .center-container {
+    
+    .upload-section {
         display: flex;
         flex-direction: column;
         align-items: center;
         justify-content: center;
-        height: 100%;
+        min-height: 50vh;
+        animation: fadeIn 0.5s ease-in;
+    }
+    
+    .results-section {
+        animation: slideUp 0.5s ease-out;
+    }
+    
+    .detection-box {
+        background: white;
+        border-radius: 8px;
+        padding: 1rem;
+        box-shadow: 0 1px 3px rgba(0,0,0,0.1);
+        margin-bottom: 1rem;
+        transform-origin: top;
+        animation: scaleIn 0.3s ease-out;
+    }
+    
+    .result-item {
+        padding: 0.5rem;
+        border-radius: 4px;
+        margin: 0.25rem 0;
+        background: #f1f5f9;
+        animation: fadeIn 0.3s ease-out;
+    }
+    
+    .image-grid {
+        display: grid;
+        grid-template-columns: repeat(auto-fill, minmax(200px, 1fr));
+        gap: 1rem;
+        margin-top: 1rem;
+    }
+    
+    .image-container {
+        background: white;
+        border-radius: 8px;
+        padding: 0.5rem;
+        box-shadow: 0 1px 3px rgba(0,0,0,0.1);
+        animation: scaleIn 0.3s ease-out;
+    }
+    
+    @keyframes fadeIn {
+        from { opacity: 0; }
+        to { opacity: 1; }
+    }
+    
+    @keyframes slideUp {
+        from { transform: translateY(20px); opacity: 0; }
+        to { transform: translateY(0); opacity: 1; }
+    }
+    
+    @keyframes scaleIn {
+        from { transform: scale(0.95); opacity: 0; }
+        to { transform: scale(1); opacity: 1; }
+    }
+    
+    /* Compact image style */
+    .stImage > img {
+        max-height: 300px !important;
+        width: auto !important;
+        margin: 0 auto;
+        object-fit: contain;
+    }
+    
+    #MainMenu, footer, header {
+        display: none !important;
     }
 </style>
 """, unsafe_allow_html=True)
@@ -63,101 +117,104 @@ def translate_label(label):
     }
     return translations.get(label.lower(), label)
 
-def draw_boxes(image, predictions):
+def draw_boxes(image: Image, predictions: List[Dict]) -> Image:
     draw = ImageDraw.Draw(image)
     for pred in predictions:
         box = pred['box']
         label = f"{translate_label(pred['label'])} ({pred['score']:.2%})"
         color = "#2563eb" if pred['score'] > 0.7 else "#eab308"
-
+        
         draw.rectangle(
             [(box['xmin'], box['ymin']), (box['xmax'], box['ymax'])],
             outline=color,
             width=2
         )
-
-        # Ajouter des points de "chaleur" aux fractures détectées
-        center_x = (box['xmin'] + box['xmax']) / 2
-        center_y = (box['ymin'] + box['ymax']) / 2
-        radius = 5
-        draw.ellipse(
-            [(center_x - radius, center_y - radius), (center_x + radius, center_y + radius)],
-            fill=color
-        )
-
-        # Label plus compact
-        draw.text((box['xmin'], box['ymin'] - 15), label, fill="white")
+        
+        text_bbox = draw.textbbox((box['xmin'], box['ymin']-15), label)
+        draw.rectangle(text_bbox, fill=color)
+        draw.text((box['xmin'], box['ymin']-15), label, fill="white")
     return image
 
 def main():
     models = load_models()
-
-    if "uploaded" not in st.session_state:
-        st.session_state["uploaded"] = False
-
-    if not st.session_state["uploaded"]:
-        st.markdown("""
-        <div class="center-container">
-            <h2>📤 Röntgenbild Hochladen</h2>
-            <p>Bitte laden Sie ein Röntgenbild hoch, um die Analyse zu starten.</p>
-        </div>
-        """, unsafe_allow_html=True)
-        uploaded_file = st.file_uploader("Röntgenbild auswählen", type=['png', 'jpg', 'jpeg'], label_visibility="collapsed")
-
-        if uploaded_file:
-            st.session_state["uploaded"] = True
-            st.session_state["file"] = uploaded_file
-            st.session_state["analyze"] = False
-    else:
-        uploaded_file = st.session_state["file"]
-
-        if not st.session_state.get("analyze", False):
-            if st.button("🔍 Analyse starten"):
-                st.session_state["analyze"] = True
-
-        if st.session_state["analyze"]:
-            col1, col2, col3 = st.columns([1, 1.5, 1])
-
-            with col1:
-                st.markdown("### 🎯 KI-Analyse")
-
-                st.markdown("**🛡️ Der KnochenWächter**")
-                image = Image.open(uploaded_file)
-                predictions_wachter = models["KnochenWächter"](image)
-                for pred in predictions_wachter:
-                    score_color = "#22c55e" if pred['score'] > 0.7 else "#eab308"
-                    st.markdown(f"""
-                        <div class='result-box'>
-                            <span style='color: {score_color}; font-weight: 500;'>
-                                {pred['score']:.1%}
-                            </span> - {translate_label(pred['label'])}
-                        </div>
-                    """, unsafe_allow_html=True)
-
-                st.markdown("**🎓 Der RöntgenMeister**")
-                predictions_meister = models["RöntgenMeister"](image)
-                for pred in predictions_meister:
-                    score_color = "#22c55e" if pred['score'] > 0.7 else "#eab308"
-                    st.markdown(f"""
-                        <div class='result-box'>
-                            <span style='color: {score_color}; font-weight: 500;'>
-                                {pred['score']:.1%}
-                            </span> - {translate_label(pred['label'])}
-                        </div>
-                    """, unsafe_allow_html=True)
-
-            with col2:
-                st.image(image, use_container_width=True)
-
-            predictions_auge = models["KnochenAuge"](image)
-            filtered_preds = [p for p in predictions_auge if p['score'] >= 0.6]
-
-            if filtered_preds:
+    
+    if 'analyzed_images' not in st.session_state:
+        st.session_state.analyzed_images = []
+    
+    # Section upload centrée
+    st.markdown('<div class="upload-section">', unsafe_allow_html=True)
+    st.markdown("### 📤 Röntgenbild Upload")
+    uploaded_files = st.file_uploader("", type=['png', 'jpg', 'jpeg'], accept_multiple_files=True)
+    conf_threshold = st.slider(
+        "Konfidenzschwelle",
+        min_value=0.0, max_value=1.0,
+        value=0.60, step=0.05
+    )
+    analyze_button = st.button("Analysieren")
+    st.markdown('</div>', unsafe_allow_html=True)
+    
+    if analyze_button and uploaded_files:
+        st.markdown('<div class="results-section">', unsafe_allow_html=True)
+        
+        for uploaded_file in uploaded_files:
+            image = Image.open(uploaded_file)
+            
+            # Animation de chargement
+            with st.spinner("Analyse läuft..."):
+                time.sleep(0.5)  # Animation effect
+                
+                col1, col2, col3 = st.columns([1, 1, 1])
+                
+                with col1:
+                    st.markdown("### 📋 Bild Details")
+                    st.image(image, use_column_width=True)
+                
+                with col2:
+                    st.markdown("### 🎯 KI-Analyse")
+                    
+                    # KnochenWächter
+                    with st.container():
+                        st.markdown("#### 🛡️ KnochenWächter")
+                        predictions = models["KnochenWächter"](image)
+                        for pred in predictions:
+                            if pred['score'] >= conf_threshold:
+                                st.markdown(f"""
+                                    <div class="result-item">
+                                        <span style='color: {"#22c55e" if pred["score"] > 0.7 else "#eab308"}; font-weight: 500;'>
+                                            {pred['score']:.1%}
+                                        </span> - {translate_label(pred['label'])}
+                                    </div>
+                                """, unsafe_allow_html=True)
+                    
+                    # RöntgenMeister
+                    with st.container():
+                        st.markdown("#### 🎓 RöntgenMeister")
+                        predictions = models["RöntgenMeister"](image)
+                        for pred in predictions:
+                            if pred['score'] >= conf_threshold:
+                                st.markdown(f"""
+                                    <div class="result-item">
+                                        <span style='color: {"#22c55e" if pred["score"] > 0.7 else "#eab308"}; font-weight: 500;'>
+                                            {pred['score']:.1%}
+                                        </span> - {translate_label(pred['label'])}
+                                    </div>
+                                """, unsafe_allow_html=True)
+                
+                # Afficher la localisation uniquement si une fracture est détectée
                 with col3:
-                    st.markdown("### 👁️ Das KnochenAuge - Lokalisation")
-                    result_image = image.copy()
-                    result_image = draw_boxes(result_image, filtered_preds)
-                    st.image(result_image, use_container_width=True)
+                    predictions_location = models["KnochenAuge"](image)
+                    fractures_detected = any(p['score'] >= conf_threshold and 'fracture' in p['label'].lower() 
+                                          for p in predictions_location)
+                    
+                    if fractures_detected:
+                        st.markdown("### 🔍 Fraktur Lokalisation")
+                        filtered_preds = [p for p in predictions_location if p['score'] >= conf_threshold]
+                        if filtered_preds:
+                            result_image = image.copy()
+                            result_image = draw_boxes(result_image, filtered_preds)
+                            st.image(result_image, use_column_width=True)
+        
+        st.markdown('</div>', unsafe_allow_html=True)
 
 if __name__ == "__main__":
-    main()
+    main()