Update api_server.py

api_server.py

@@ -74,7 +74,21 @@ def predict():
     # 檢查 YOLO 是否返回了有效的結果
     if results is None or len(results) == 0:
         return jsonify({'error': 'No results from YOLO model'}), 400
-        
+
+    # 渲染推理結果到圖像
+    img_with_boxes = results[0].plot()  # 使用 results[0]，假設只有一張圖像作推理
+
+    # 將 numpy array 轉換為 PIL Image
+    img = Image.fromarray(img_with_boxes)
+
+    # 儲存圖片到內存緩衝區
+    img_io = io.BytesIO()
+    img.save(img_io, 'PNG')
+    img_io.seek(0)
+
+    # 返回處理後的圖像
+    return send_file(img_io, mimetype='image/png')
+  '''      
     saved_images = []
     
     # 儲存辨識後的圖片到指定資料夾
@@ -100,31 +114,9 @@ def predict():
         'saved_images': saved_images,
         'inference_time': inference_time
     }), 200
+'''
 
 
-
-    # # Process the YOLO output
-    # detections = []
-    # for det in results.xyxy[0]:  # Assuming results are in xyxy format (xmin, ymin, xmax, ymax, confidence, class)
-    #     x_min, y_min, x_max, y_max, confidence, class_idx = det
-    #     width = x_max - x_min
-    #     height = y_max - y_min
-    #     detection = {
-    #         "label": int(class_idx),
-    #         "confidence": float(confidence),
-    #         "bbox": [float(x_min), float(y_min), float(width), float(height)]
-    #     }
-    #     detections.append(detection)
-
-    # # Calculate latency in milliseconds
-    # latency_ms = (time.time() - start_time) * 1000
-
-    # # Return the detection results and latency as JSON response
-    # response = {
-    #     'detections': detections,
-    #     'ml-latency-ms': round(latency_ms, 4)
-    # }
-
     # # dictionary is not a JSON: https://www.quora.com/What-is-the-difference-between-JSON-and-a-dictionary
     # # flask.jsonify vs json.dumps https://sentry.io/answers/difference-between-json-dumps-and-flask-jsonify/
     # # The flask.jsonify() function returns a Response object with Serializable JSON and content_type=application/json.