Update app.py

app.py

@@ -27,82 +27,85 @@ def detect_objects(image):
         temp_file_path = temp_file.name
 
     try:
-        # Perform sliced inference with SAHI
-        result = get_sliced_prediction(
-            temp_file_path,
-            model,
-            slice_height=256,  # Adjust slice height as needed
-            slice_width=256,   # Adjust slice width as needed
-            overlap_height_ratio=0.2,  # Adjust overlap height ratio as needed
-            overlap_width_ratio=0.2   # Adjust overlap width ratio as needed
-        )
-
-        # Menghitung jumlah objek per kelas
-        class_count = {}
-        total_count = 0  # Menyimpan total jumlah objek
-
-        for prediction in result.object_prediction_list:
-            class_name = prediction.class_id  # or prediction.class_name if available
-            class_count[class_name] = class_count.get(class_name, 0) + 1
-            total_count += 1  # Tambah jumlah objek untuk setiap prediksi
+        # Perform sliced inference with SAHI using InferenceSlicer
+        def callback(image_slice: np.ndarray) -> sv.Detections:
+            results = model.infer(image_slice)[0]  # Perform inference on each slice
+            return sv.Detections.from_inference(results)
 
-        # Menyusun output berupa string hasil perhitungan
-        result_text = "Product Nestle\n\n"
-        for class_name, count in class_count.items():
-            result_text += f"{class_name}: {count}\n"
-        result_text += f"\nTotal Product Nestle: {total_count}"
+        # Configure the SAHI Slicer with specific slice dimensions and overlap
+        slicer = sv.InferenceSlicer(
+            callback=callback,
+            slice_wh=(320, 320),  # Adjust slice dimensions as needed
+            overlap_wh=(0.2, 0.2),  # Adjust overlap ratio for better results
+            overlap_filter=sv.OverlapFilter.NON_MAX_SUPPRESSION,  # Filter overlapping detections
+            iou_threshold=0.5,  # Intersection over Union threshold for NMS
+        )
 
-        # Menyimpan gambar dengan prediksi
-        output_image_path = "/tmp/prediction.jpg"
-        result.export_visuals(export_dir="/tmp/")  # Export visuals for display
-        output_image_path = "/tmp/prediction_visual.png"  # Assuming the visual output is saved here
+        # Run slicing-based inference
+        detections = slicer(image)
 
-        # Annotating the image with the detections (optional)
-        label_annotator = sv.LabelAnnotator()
+        # Annotate the results on the image
         box_annotator = sv.BoxAnnotator()
+        label_annotator = sv.LabelAnnotator()
 
         annotated_image = box_annotator.annotate(
-            scene=image.copy(), detections=result.object_prediction_list)
+            scene=image.copy(), detections=detections)
         
         annotated_image = label_annotator.annotate(
-            scene=annotated_image, detections=result.object_prediction_list)
+            scene=annotated_image, detections=detections)
 
         # Save the annotated image
+        output_image_path = "/tmp/prediction_visual.png"
         annotated_image.save(output_image_path)
 
+        # Count the number of detected objects per class
+        class_count = {}
+        total_count = 0
+
+        for prediction in detections:
+            class_name = prediction.class_id  # or prediction.class_name if available
+            class_count[class_name] = class_count.get(class_name, 0) + 1
+            total_count += 1  # Increment the total object count
+
+        # Create a result text with object counts
+        result_text = "Detected Objects:\n\n"
+        for class_name, count in class_count.items():
+            result_text += f"{class_name}: {count}\n"
+        result_text += f"\nTotal objects detected: {total_count}"
+
     except requests.exceptions.HTTPError as http_err:
-        # Menangani kesalahan HTTP
+        # Handle HTTP errors
         result_text = f"HTTP error occurred: {http_err}"
-        output_image_path = temp_file_path  # Kembalikan gambar asli jika terjadi error
+        output_image_path = temp_file_path  # Return the original image in case of error
     except Exception as err:
-        # Menangani kesalahan lain
+        # Handle other errors
         result_text = f"An error occurred: {err}"
-        output_image_path = temp_file_path  # Kembalikan gambar asli jika terjadi error
+        output_image_path = temp_file_path  # Return the original image in case of error
 
-    # Hapus file sementara setelah prediksi
+    # Clean up temporary files
     os.remove(temp_file_path)
     
     return output_image_path, result_text
 
-# Membuat antarmuka Gradio dengan tata letak fleksibel
+# Create the Gradio interface
 with gr.Blocks() as iface:
     with gr.Row():
         with gr.Column():
             input_image = gr.Image(type="pil", label="Input Image")
         with gr.Column():
-            output_image = gr.Image(label="Detect Object")
+            output_image = gr.Image(label="Detected Objects")
         with gr.Column():
-            output_text = gr.Textbox(label="Counting Object")
+            output_text = gr.Textbox(label="Object Count")
     
-    # Tombol untuk memproses input
-    detect_button = gr.Button("Detect")
+    # Button to trigger object detection
+    detect_button = gr.Button("Detect Objects")
     
-    # Hubungkan tombol dengan fungsi deteksi
+    # Link the button to the detect_objects function
     detect_button.click(
         fn=detect_objects, 
         inputs=input_image, 
         outputs=[output_image, output_text]
     )
 
-# Menjalankan antarmuka
+# Launch the interface
 iface.launch()