Model Prediction

1. CLI Usage

CLI currently only supports calling USB cameras. For CSI cameras, users can directly modify the previous Python code to call the ncnn model!

cd ~/ultralytics/ultralytics/data/yahboom_data/orange_data/runs/detect/train2/weights
yolo predict model='best.engine' source=0 save=False show   # For multiple cameras, change the number after source

2. Python Usage

Use Python to call USB cameras and CSI cameras to recognize oranges.

2.1. USB Camera

Use best_ncnn_model to predict camera footage:

xxxxxxxxxx
cd /home/jetson/ultralytics/yahboom_demo

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

xxxxxxxxxx
python3 06.orange_camera_usb.py

Effect Preview

yolo recognition output video location: /home/jetson/ultralytics/output/

Example code:

​x
import cv2
from ultralytics import YOLO
​
model = YOLO("/home/jetson/ultralytics/ultralytics/data/yahboom_data/orange_data/runs/detect/train2/weights/best.engine")
​
# Open the cammera
cap = cv2.VideoCapture(0)
cap.set(6, cv2.VideoWriter.fourcc('M', 'J', 'P', 'G'))
cap.set(cv2.CAP_PROP_FRAME_WIDTH, 640)
cap.set(cv2.CAP_PROP_FRAME_HEIGHT, 480)
​
# 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/06.orange_camera_usb.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", cv2.resize(annotated_frame, (640, 480)))
​
        # 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()
​

2.2. CSI Camera

Use best_ncnn_model to predict camera footage:

xxxxxxxxxx
cd /home/jetson/ultralytics/yahboom_demo

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

xxxxxxxxxx
python3 06.orange_camera_csi.py

Effect Preview

yolo recognition output video location: /home/jetson/ultralytics/output/

Example code:

xxxxxxxxxx
import cv2
from ultralytics import YOLO
from jetcam.csi_camera import CSICamera
​
# Load the YOLO model
model = YOLO("/home/jetson/ultralytics/ultralytics/data/yahboom_data/orange_data/runs/detect/train2/weights/best.engine")
​
# Open the camera (CSI Camera)
cap = CSICamera(capture_device=0, width=640, height=480)
​
# Get the video frame size and frame rate
frame_width = 640
frame_height = 480
fps = 30
​
# Define the codec and create a VideoWriter object to output the processed video
output_path = "/home/jetson/ultralytics/output/06.orange_camera_csi.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 True:
    # Read a frame from the camera
    frame = cap.read()
​
    if frame is not None:
        # 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", cv2.resize(annotated_frame, (640, 480)))
​
        # Break the loop if 'q' is pressed
        if cv2.waitKey(1) & 0xFF == ord("q"):
            break
    else:
        # Break the loop if no frame is received (camera error or end of stream)
        print("No frame received, breaking the loop.")
        break
​
# Release the video capture and writer objects, and close the display window
cap.release()
out.release()
cv2.destroyAllWindows()
​