Update app.py

app.py

@@ -7,12 +7,12 @@ import numpy as np
 import cv2
 from PIL import Image, ImageDraw
 
-# Inisialisasi Roboflow (for model path)
+# Initialize Roboflow
 rf = Roboflow(api_key="Otg64Ra6wNOgDyjuhMYU")
 project = rf.workspace("alat-pelindung-diri").project("nescafe-4base")
 model = project.version(16).model
 
-# Fungsi untuk melakukan Non-Maximum Suppression (NMS)
+# Apply NMS (Non-Maximum Suppression)
 def apply_nms(predictions, iou_threshold=0.5):
     boxes = []
     scores = []
@@ -31,6 +31,9 @@ def apply_nms(predictions, iou_threshold=0.5):
     # Perform NMS using OpenCV
     indices = cv2.dnn.NMSBoxes(boxes.tolist(), scores.tolist(), score_threshold=0.25, nms_threshold=iou_threshold)
 
+    print(f"Predictions before NMS: {predictions}")
+    print(f"Indices after NMS: {indices}")
+
     # Check if indices is empty or invalid
     if not indices or not isinstance(indices, tuple) or len(indices) == 0:
         print("No valid indices returned from NMS.")
@@ -53,14 +56,14 @@ def apply_nms(predictions, iou_threshold=0.5):
 
     return nms_predictions
 
-# Fungsi untuk deteksi objek menggunakan Roboflow Model
+# Detect objects and annotate the image
 def detect_objects(image):
-    # Menyimpan gambar sementara
+    # Save the image temporarily
     with tempfile.NamedTemporaryFile(delete=False, suffix=".jpg") as temp_file:
         image.save(temp_file, format="JPEG")
         temp_file_path = temp_file.name
 
-    # Slice gambar menjadi potongan-potongan kecil
+    # Slice the image into smaller pieces
     slice_image_result = slice_image(
         image=temp_file_path,
         output_file_name="sliced_image",
@@ -82,10 +85,10 @@ def detect_objects(image):
         print("Failed to access sliced_image_paths attribute.")
         sliced_image_paths = []
 
-    # Menyimpan semua prediksi untuk setiap potongan gambar
+    # Save all predictions for each sliced image
     all_predictions = []
 
-    # Prediksi pada setiap potongan gambar
+    # Predict on each sliced image
     for sliced_image_path in sliced_image_paths:
         if isinstance(sliced_image_path, str):
             predictions = model.predict(image_path=sliced_image_path).json()
@@ -93,10 +96,10 @@ def detect_objects(image):
         else:
             print(f"Skipping invalid image path: {sliced_image_path}")
     
-    # Aplikasikan NMS untuk menghapus duplikat deteksi
+    # Apply NMS to remove duplicate detections
     postprocessed_predictions = apply_nms(all_predictions, iou_threshold=0.5)
 
-    # Annotate gambar dengan hasil prediksi menggunakan OpenCV
+    # Annotate the image with prediction results using OpenCV
     img = cv2.imread(temp_file_path)
     for prediction in postprocessed_predictions:
         class_name = prediction['class']
@@ -118,11 +121,11 @@ def detect_objects(image):
     img_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
     annotated_image = Image.fromarray(img_rgb)
 
-    # Simpan gambar hasil annotasi
+    # Save the annotated image
     output_image_path = "/tmp/prediction.jpg"
     annotated_image.save(output_image_path)
 
-    # Menghitung jumlah objek per kelas
+    # Count objects per class
     class_count = {}
     for detection in postprocessed_predictions:
         class_name = detection['class']
@@ -131,23 +134,23 @@ def detect_objects(image):
         else:
             class_count[class_name] = 1
 
-    # Hasil perhitungan objek
+    # Object count result
     result_text = "Jumlah objek per kelas:\n"
     for class_name, count in class_count.items():
         result_text += f"{class_name}: {count} objek\n"
 
-    # Hapus file sementara
+    # Remove temporary file
     os.remove(temp_file_path)
 
     return output_image_path, result_text
 
-# Membuat antarmuka Gradio
+# Gradio interface
 iface = gr.Interface(
-    fn=detect_objects,                         # Fungsi yang dipanggil saat gambar diupload
-    inputs=gr.Image(type="pil"),               # Input berupa gambar
-    outputs=[gr.Image(), gr.Textbox()],        # Output gambar dan teks
-    live=True                                    # Menampilkan hasil secara langsung
+    fn=detect_objects,                         # Function called when image is uploaded
+    inputs=gr.Image(type="pil"),               # Input is an image
+    outputs=[gr.Image(), gr.Textbox()],        # Output is an image and text
+    live=True                                    # Display results live
 )
 
-# Menjalankan antarmuka
+# Run the interface
 iface.launch()