| 5 | device = torch.device('cuda' if torch.cuda.is_available() else 'cpu') |
| 6 | |
| 7 | def main(): |
| 8 | |
| 9 | # Load YOLOv9 model |
| 10 | model = YOLO('yolov9c.pt') |
| 11 | model.to(device) |
| 12 | |
| 13 | # Initialize video capture (0 for default camera) |
| 14 | cap = cv2.VideoCapture(0) |
| 15 | |
| 16 | # Check if the video capture device is opened |
| 17 | if not cap.isOpened(): |
| 18 | print("Error: Could not open video capture device") |
| 19 | return |
| 20 | |
| 21 | while True: |
| 22 | # Capture frame-by-frame |
| 23 | ret, frame = cap.read() |
| 24 | |
| 25 | if not ret: |
| 26 | print("Error: Could not read frame") |
| 27 | break |
| 28 | |
| 29 | # Use YOLOv9 model to make predictions |
| 30 | results = model(frame) |
| 31 | |
| 32 | # Process the results |
| 33 | for result in results: |
| 34 | # Loop through each detected object |
| 35 | for box in result.boxes: |
| 36 | # Get coordinates and class label |
| 37 | x1, y1, x2, y2 = box.xyxy[0] |
| 38 | label_id = int(box.cls[0].item()) |
| 39 | confidence = box.conf[0].item() |
| 40 | |
| 41 | # Get the class label from YOLO model |
| 42 | class_label = model.names[label_id] |
| 43 | |
| 44 | # Create the label text |
| 45 | label_text = f"{class_label}: {confidence:.2f}" |
| 46 | |
| 47 | # Draw bounding box on the frame |
| 48 | cv2.rectangle(frame, (int(x1), int(y1)), (int(x2), int(y2)), (0, 255, 0), 2) |
| 49 | |
| 50 | # Draw the label text on the frame above the bounding box |
| 51 | cv2.putText(frame, label_text, (int(x1), int(y1) - 5), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 1) |
| 52 | |
| 53 | # Display the frame with bounding boxes and labels |
| 54 | cv2.imshow('Real-Time Object Detection', frame) |
| 55 | |
| 56 | # Exit the loop if the user presses 'q' |
| 57 | if cv2.waitKey(1) & 0xFF == ord('q'): |
| 58 | break |
| 59 | |
| 60 | # Release the video capture device and close the window |
| 61 | cap.release() |
| 62 | cv2.destroyAllWindows() |
| 63 | |
| 64 | if __name__ == "__main__": |