Update app.py

app.py

@@ -15,14 +15,11 @@ st.set_page_config(page_title="Car Logo AI Pro 2026", layout="wide", page_icon="
 @st.cache_resource
 def load_car_dataset(base_path):
     dataset = []
-    if not os.path.exists(base_path):
-        image_files = [f for f in os.listdir('.') if f.lower().endswith(('.png', '.jpg', '.jpeg'))]
-        for f in image_files:
-            img = cv2.imread(f, 0)
-            if img is not None:
-                dataset.append((f, cv2.resize(img, (100, 100))))
-    else:
-        for root, dirs, files in os.walk(base_path):
+    # Klasör yolunu normalize et (Windows/Linux uyumu için)
+    target_path = os.path.normpath(base_path)
+    
+    if os.path.exists(target_path):
+        for root, dirs, files in os.walk(target_path):
             for file in files:
                 if file.lower().endswith(('.png', '.jpg', '.jpeg')):
                     full_path = os.path.join(root, file)
@@ -31,6 +28,7 @@ def load_car_dataset(base_path):
                         dataset.append((full_path, cv2.resize(img, (100, 100))))
     return dataset
 
+# Veri setini yükle
 dataset = load_car_dataset('Car_Logo_Dataset')
 
 # --- 4. RESİM İŞLEME / IMAGE PROCESSING ---
@@ -52,9 +50,9 @@ col_left, col_right = st.columns([1, 1], gap="large")
 with col_left:
     st.subheader("📤 Yükleme Alanı / Upload Area")
     
-    # İSTEDİĞİN ÖZEL YAZIYI BURAYA EKLEDİK
+    # İstediğin özel 200MB uyarısı
     st.markdown("""
-        <div style="color: #555; font-size: 0.9em; margin-bottom: -15px;">
+        <div style="color: #666; font-size: 0.85em; margin-bottom: -10px; font-family: sans-serif;">
             200MB per file • JPG, PNG, JPEG
         </div>
     """, unsafe_allow_html=True)
@@ -62,8 +60,7 @@ with col_left:
     uploaded_file = st.file_uploader(
         "Yeni bir logo seçin / Select a new logo", 
         type=['jpg', 'png', 'jpeg'],
-        key="main_uploader",
-        label_visibility="visible"
+        key="main_uploader"
     )
     
     st.markdown("---")
@@ -75,11 +72,11 @@ with col_left:
 
 with col_right:
     st.subheader("🎯 Analiz ve Sonuç / Analysis & Result")
-    threshold = st.slider("Hassasiyet / Sensitivity", 0.0, 1.0, 0.20)
+    threshold = st.slider("Hassasiyet / Sensitivity (Threshold)", 0.0, 1.0, 0.20)
     
     if uploaded_file:
         if st.button("ŞİMDİ TANI / PREDICT NOW", type="primary", use_container_width=True):
-            with st.spinner("32 Marka Taranıyor... / Scanning..."):
+            with st.spinner("32 Marka Taranıyor... / Scanning 32 Brands..."):
                 query = cv2.resize(gray_img, (100, 100))
                 best_score = -1
                 best_path = None
@@ -91,34 +88,42 @@ with col_right:
                         best_score = max_val
                         best_path = path
                 
+                # --- AKILLI MARKA İSMİ AYIKLAMA ---
                 if best_path and best_score >= threshold:
                     st.balloons()
-                    folder_name = os.path.basename(os.path.dirname(best_path))
-                    if not folder_name or folder_name == '.':
-                        folder_name = os.path.basename(best_path).split('.')[0]
                     
-                    brand_name = ''.join([i for i in folder_name if not i.isdigit() and i not in ['-', '_']]).strip().upper()
+                    # Yolu parçalarına ayır
+                    parts = os.path.normpath(best_path).split(os.sep)
+                    
+                    # Klasör yapısına göre marka adını bul (Genelde sondan bir önceki parça)
+                    if len(parts) > 1:
+                        raw_brand = parts[-2]
+                        # Eğer üst klasör ismini (Car_Logo_Dataset) aldıysa dosya adına bak
+                        if "DATASET" in raw_brand.upper() or raw_brand == ".":
+                            raw_brand = parts[-1].split('.')[0]
+                    else:
+                        raw_brand = parts[-1].split('.')[0]
+
+                    # Gereksiz karakterleri ve sayıları temizle
+                    brand_name = ''.join([i for i in raw_brand if not i.isdigit() and i not in ['-', '_']]).strip().upper()
                     
                     st.success(f"### TAHMİN / PREDICTION: **{brand_name}**")
                     st.metric("Benzerlik / Similarity", f"%{int(best_score*100)}")
-                    st.image(best_path, width=150, caption=f"Eşleşen Referans: {brand_name}")
+                    st.image(best_path, width=150, caption=f"Eşleşen: {brand_name}")
                 else:
                     st.error("❌ Eşleşme Bulunamadı / Match Not Found")
-    else:
-        st.info("Lütfen soldaki panelden bir logo yükleyerek başlayın.")
 
 # --- 6. TEKNİK ANALİZ ---
 if uploaded_file:
     st.markdown("---")
     with st.expander("🔍 Teknik Detaylar / Technical Details"):
-        tab_tr, tab_en = st.tabs(["🇹🇷 Türkçe", "🇺🇸 English"])
-        with tab_tr:
-            st.write("### Görüntü İşleme Aşamaları")
-            c1, c2 = st.columns(2)
-            c1.image(edge_img, width=300, caption="Canny Kenar Analizi")
-            c2.image(cv2.resize(gray_img, (100, 100)), caption="Sistemin Gördüğü (100x100)")
-        with tab_en:
-            st.write("### Image Processing Stages")
-            ce1, ce2 = st.columns(2)
-            ce1.image(edge_img, width=300, caption="Canny Edge Detection")
-            ce2.image(cv2.resize(gray_img, (100, 100)), caption="AI View (100x100)")
+        t1, t2 = st.tabs(["🇹🇷 Türkçe", "🇺🇸 English"])
+        with t1:
+            st.image(edge_img, width=300, caption="Kenar Analizi")
+            st.info(f"Sistem veri setindeki {len(dataset)} referans ile karşılaştırma yaptı.")
+        with t2:
+            st.image(edge_img, width=300, caption="Edge Analysis")
+            st.info(f"System compared with {len(dataset)} reference images.")
+
+# Görsel Stil
+st.markdown("<style>.stMetric { background: #f0f2f6; border-radius: 10px; padding: 10px; }</style>", unsafe_allow_html=True)