Raspberry Pi YOLO Object Detection Tutorial (Beginner Friendly)

Add YOLO to Your Existing Raspberry Pi Virtual Environment.

What You Will Need

Hardware

• Raspberry Pi 3, 4, or 5.
• USB Camera.

Software

Before starting this tutorial, you should already have:

• OpenCV installed.
• a virtual environment set up

If not already installed, follow the CraftyRobotics tutorial on Virtual Environments.

Step 1 — Activate Your Virtual Environment

We will be using the CraftyRobotics virtual environment created in the CraftyRobotics tutorial on virtual environments.

From the Raspberry Pi desktop, open a terminal.

Update Raspberry Pi:

In terminal enter:

sudo apt update
sudo apt upgrade -y

This will take a few minutes to complete.

Activate the Virtual Environment:

Activate the Virtual Environment by entering.

source crafty_env/bin/activate

If you are using a different virtual environment, replace "crafty_env" with the name of your virtual environment.

Your terminal will now look like this:

(crafty_env) pi@raspberrypi:~ $

That (crafty_env) at the beginning means you are inside the virtual environment.

Install YOLO

In the terminal, enter:

pip install ultralytics

This may take some time on Raspberry Pi. Almost an hour on my system.

Ultralytics a powerful system that makes machine vision and object detection much easier to use with Python.

Once Ultralytics has completed close the terminal.

Step 2 — Create a new folder for our program

Create a new folder for the YOLO project files..

This helps keep:

• Python files
• models
• images
• project files

organized in one location.

Click on the desktop, then right-click and select “New Folder”. Name the new folder “AI_Vision”.

Step 3 — Open Thonny

Launch Thonny from the Raspberry Pi menu.

Step 4 — Check Thonny Interpreter

Check to make sure Thonny is connected to the virtual environment.

You should see the path to the virtual environment and the interpreter (python3) in the lower-left corner of the Thonny window. If not, follow the CraftyRobotics tutorial on virtual environments.

/home/pi/crafty_env/python3

Step 5 — Create the YOLO Program

Copy the code below and paste it into the Thonny editor (upper window). A description of the code will be provided at the end of the tutorial.

from ultralytics import YOLO
import cv2

# Load YOLO model
model = YOLO("yolov8n.pt")

# Open webcam
cap = cv2.VideoCapture(0)

while True:
    ret, frame = cap.read()

    if not ret:
        break

    # Run object detection
    results = model(frame)

    # Draw detection boxes
    annotated_frame = results[0].plot()

    # Show video
    cv2.imshow("YOLO Detection", annotated_frame)

    # Press Q to quit
    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

cap.release()
cv2.destroyAllWindows()

Step 6 — Save the Program

Click Save, navigate to the AI_Vision folder, and name the file yolo_test.py and click enter.

yolo_test.py

Step 7 — Run the Program

Click the green Run button in Thonny.

The first launch downloads the YOLO model automatically.

After that:

• the webcam opens
• objects are detected
• labels appear on screen in real time

Code Description

This program uses YOLO (You Only Look Once) and OpenCV to perform real-time object detection using a USB camera connected to the Raspberry Pi.

The program starts by importing:

• YOLO from Ultralytics for AI object detection
• cv2 from OpenCV for camera access and video display

from ultralytics import YOLO
import cv2

Next, the YOLO model is loaded.

model = YOLO("yolov8n.pt")

The yolov8n.pt model is a lightweight version of YOLO designed to run efficiently on systems like the Raspberry Pi.

The webcam is then opened using OpenCV.

cap = cv2.VideoCapture(0)

0 tells OpenCV to use the default USB camera.

The program then enters a continuous loop where it:

• captures a video frame
• sends the frame to YOLO
• detects objects
• draws detection boxes and labels
• displays the updated video feed

results = model(frame)

YOLO analyzes the image and identifies objects such as:

• people
• bottles
• keyboards
• cups
• chairs

The detected objects are then drawn onto the frame.

annotated_frame = results[0].plot()

This automatically adds:

• bounding boxes
• labels
• confidence scores

to the video feed.

The updated frame is displayed in a window titled:

YOLO Detection

Finally, the program checks if the Q key is pressed.

if cv2.waitKey(1) & 0xFF == ord('q'):

If Q is pressed:

• the loop stops
• the camera closes
• all OpenCV windows are cleaned up safely

This creates a simple but powerful real-time AI vision system on Raspberry Pi using Python, OpenCV, and YOLO.

What’s Next?

• AI robot vision
• Object tracking
• Smart security cameras
• Motion-triggered AI detection
• Face detection
• Autonomous navigation
• Garden robotics
• Real-time image capture

You can also experiment with:

• different YOLO models
• faster camera settings
• custom-trained AI models
• multiple cameras

Back to Craftyrobotics.com.

Raspberry Pi Tutorials

Craftyrobotics.com/