Object detection

Using Python to demonstrate Ultralytics: Object Detection in images, videos, and real-time detection.

1. Enable optimal performance of the board

Enable Jetson Clocks

Enabling Jetson Clocks will ensure that all CPU and GPU cores run at maximum frequency:

sudo jetson_clocks

2. Target prediction: image

Use yolo11n.pt to predict the images provided by the ultralytics project.

Enter the code folder:

xxxxxxxxxx
cd /home/jetson/ultralytics/ultralytics/yahboom_demo

Run the code:

xxxxxxxxxx
python3 01.detection_image.py

Effect preview

Yolo recognition output image location: /home/jetson/ultralytics/ultralytics/output/

Sample code:

​x
from ultralytics import YOLO
​
# Load a model
model = YOLO("/home/jetson/ultralytics/ultralytics/yolo11n.pt")
​
# Run batched inference on a list of images
results = model("/home/jetson/ultralytics/ultralytics/assets/bus.jpg")  # return a list of Results objects
​
# Process results list
for result in results:
    boxes = result.boxes  # Boxes object for bounding box outputs
    # masks = result.masks  # Masks object for segmentation masks outputs
    # keypoints = result.keypoints  # Keypoints object for pose outputs
    # probs = result.probs  # Probs object for classification outputs
    # obb = result.obb  # Oriented boxes object for OBB outputs
    result.show()  # display to screen
    result.save(filename="/home/jetson/ultralytics/ultralytics/output/bus_output.jpg")  # save to disk

3. Target prediction: video

Use yolo11n.pt to predict videos under the ultralytics project (not the videos that come with ultralytics).

Enter the code folder:

xxxxxxxxxx
cd /home/jetson/ultralytics/ultralytics/yahboom_demo

Run the code:

xxxxxxxxxx
python3 01.detection_video.py

Effect preview

Video location of yolo recognition output: /home/jetson/ultralytics/ultralytics/output/

Sample code:

xxxxxxxxxx
import cv2
from ultralytics import YOLO
​
# Load the YOLO model
model = YOLO("/home/jetson/ultralytics/ultralytics/yolo11n.pt")
​
# Open the video file
video_path = "/home/jetson/ultralytics/ultralytics/videos/people_animals.mp4"
cap = cv2.VideoCapture(video_path)
​
# Get the video frame size and frame rate
frame_width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
frame_height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
fps = int(cap.get(cv2.CAP_PROP_FPS))
​
# Define the codec and create a VideoWriter object to output the processed video
output_path = "/home/jetson/ultralytics/ultralytics/output/01.people_animals_output.mp4"
fourcc = cv2.VideoWriter_fourcc(*'mp4v')  # You can use 'XVID' or 'mp4v' depending on your platform
out = cv2.VideoWriter(output_path, fourcc, fps, (frame_width, frame_height))
​
# Loop through the video frames
while cap.isOpened():
    # Read a frame from the video
    success, frame = cap.read()
​
    if success:
        # Run YOLO inference on the frame
        results = model(frame)
​
        # Visualize the results on the frame
        annotated_frame = results[0].plot()
​
        # Write the annotated frame to the output video file
        out.write(annotated_frame)
​
        # Display the annotated frame
        cv2.imshow("YOLO Inference", annotated_frame)
​
        # Break the loop if 'q' is pressed
        if cv2.waitKey(1) & 0xFF == ord("q"):
            break
    else:
        # Break the loop if the end of the video is reached
        break
# Release the video capture and writer objects, and close the display window
cap.release()
out.release()
cv2.destroyAllWindows()

4. Target prediction: real-time detection

Use yolo11n.engine to predict the camera image.

Start the camera.

xxxxxxxxxx
#usb camera
ros2 launch usb_cam camera.launch.py 
#nuwa camera
ros2 launch ascamera hp60c.launch.py

Enter the code folder:

xxxxxxxxxx
cd /home/jetson/ultralytics/ultralytics/yahboom_demo

Run the code: Click the preview screen and press the q key to terminate the program!

xxxxxxxxxx
python3 01.detection_camera_usb.py

Effect preview

Video location of yolo recognition output: /home/jetson/ultralytics/ultralytics/output/

Sample code:

x
#!/usr/bin/env python3
import rclpy
from rclpy.node import Node
from sensor_msgs.msg import Image, CompressedImage
from cv_bridge import CvBridge
import cv2
from ultralytics import YOLO
import os
import time
​
class Image_detection(Node):
    def __init__(self):
        super().__init__('Image_detection')
        self.model = YOLO("/home/jetson/ultralytics/ultralytics/yolo11n.engine")
        self.camera_type = os.getenv('CAMERA_TYPE', 'usb')
        self.bridge = CvBridge()
        if self.camera_type == 'usb':
            topic_name = '/usb_cam/image_raw'
        else:
            topic_name = '/ascamera_hp60c/camera_publisher/rgb0/image'
​
        self.subscription = self.create_subscription(Image,topic_name, self.image_callback,10)
​
        # Get the video frame size and frame rate
        frame_width = 640
        frame_height = 480
        fps = 30
        output_path = "/home/jetson/ultralytics/ultralytics/output/01.detection_camera_usb.mp4"
        fourcc = cv2.VideoWriter_fourcc(*'mp4v')  # You can use 'XVID' or 'mp4v' depending on your platform
        self.out = cv2.VideoWriter(output_path, fourcc, fps, (frame_width, frame_height))
        
        # Initialize time for FPS calculation
        self.prev_time = 0
​
    def image_callback(self, msg):
        cv_image = self.bridge.imgmsg_to_cv2(msg, desired_encoding='bgr8')
 
        self.proecc(cv_image)
​
            
# Loop through the video frames
    def proecc(self,frame):
        # Run YOLO inference on the frame
        results = self.model(frame)
​
        # Visualize the results on the frame
        annotated_frame = results[0].plot()
​
        # Calculate FPS
        current_time = time.time()
        fps = 1 / (current_time - self.prev_time)
        self.prev_time = current_time
        
        # Display FPS on frame
        cv2.putText(annotated_frame, f"FPS: {fps:.2f}", (10, 30), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 255, 0), 2)
​
        # Write the annotated frame to the output video file
        self.out.write(annotated_frame)
​
​
        # Display the annotated frame
        cv2.imshow("YOLO Inference", cv2.resize(annotated_frame, (640, 480)))
​
        # Break the loop if 'q' is pressed
        cv2.waitKey(1) & 0xFF == ord("q")
​
    def cancel(self):
        cv2.destroyAllWindows()
        self.out.release()
        
​
def main(args=None):
    rclpy.init(args=args)
    node = Image_detection()
    try:
        rclpy.spin(node)
    except KeyboardInterrupt:
        pass
    finally:
        node.cancel()
        node.destroy_node()
        rclpy.shutdown()
​
if __name__ == '__main__':
    main()

References

https://docs.ultralytics.com/tasks/detect/