DEV Community

Ikegbo Ogochukwu
Ikegbo Ogochukwu

Posted on

🚀 Setting Up Raspberry Pi & Camera for AI Projects (Complete Command Guide)

#ai #computervision #raspberrypi #iot

If you're working with a Raspberry Pi for computer vision or AI projects, getting everything set up can be confusing—especially with newer devices like the Raspberry Pi 5.

This guide compiles all essential commands used for:

  • Setting up your Raspberry Pi
  • Configuring network access
  • Enabling remote control
  • Troubleshooting camera issues
  • Capturing images

🧠 1. System Setup & Updates

Before doing anything, your system should be updated:

sudo apt update
sudo apt upgrade
Enter fullscreen mode Exit fullscreen mode

🌐 2. Network & IP Configuration

Check IP address 

hostname -I
Enter fullscreen mode Exit fullscreen mode

👉 This is needed for:

  • SSH connection
  • VNC connection

🔐 3. Remote Access Setup

Enable SSH and VNC 

sudo raspi-config
Enter fullscreen mode Exit fullscreen mode

Then navigate:

Interface Options → Enable SSH
Interface Options → Enable VNC
Enter fullscreen mode Exit fullscreen mode

Connect via SSH 

ssh pi@<IP_ADDRESS>
Enter fullscreen mode Exit fullscreen mode

Example:

ssh pi@192.168.43.25
Enter fullscreen mode Exit fullscreen mode

🖥️ 4. VNC (Remote Desktop Access)

Used when you want full GUI access from your laptop.

Steps:

  1. Enable VNC (see above)
  2. Install VNC Viewer on your laptop
  3. Connect using: 
<IP_ADDRESS>
Enter fullscreen mode Exit fullscreen mode

📸 5. Camera System (Important Update)

On modern Raspberry Pi OS:

Camera is enabled by default (libcamera system)
Enter fullscreen mode Exit fullscreen mode

👉 You will NOT see "Camera" in raspi-config anymore.

🧪 6. Test Camera

Preview camera 

rpicam-hello
Enter fullscreen mode Exit fullscreen mode

👉 Works only on GUI (monitor or VNC)

Capture image 

rpicam-still -o test.jpg
Enter fullscreen mode Exit fullscreen mode

Record video 

rpicam-vid -t 5000 -o video.h264
Enter fullscreen mode Exit fullscreen mode

📂 7. File Management Commands

List files 

ls
Enter fullscreen mode Exit fullscreen mode

Create folder 

mkdir dataset
cd dataset
Enter fullscreen mode Exit fullscreen mode

🔁 8. Capture Multiple Images 

for i in {1..10}; do rpicam-still -o img_$i.jpg; sleep 2; done
Enter fullscreen mode Exit fullscreen mode

🖼️ 9. Open Image (on Raspberry Pi GUI) 

xdg-open test.jpg
Enter fullscreen mode Exit fullscreen mode

🧪 10. Python + OpenCV Setup

Start Python 

python3
Enter fullscreen mode Exit fullscreen mode

Import OpenCV 

import cv2
Enter fullscreen mode Exit fullscreen mode

Load image 

img = cv2.imread("test.jpg")
Enter fullscreen mode Exit fullscreen mode

Resize image 

resized = cv2.resize(img, (224, 224))
cv2.imwrite("resized.jpg", resized)
Enter fullscreen mode Exit fullscreen mode

Convert to grayscale 

gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
cv2.imwrite("gray.jpg", gray)
Enter fullscreen mode Exit fullscreen mode

Convert to HSV 

hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
cv2.imwrite("hsv.jpg", hsv)
Enter fullscreen mode Exit fullscreen mode

Exit Python 

exit()
Enter fullscreen mode Exit fullscreen mode

⚠️ 11. Common Troubleshooting

Camera not working?

Check:

  • Cable orientation
  • Proper connection
  • Correct port (Pi 5 has 2 ports)

No preview? 

SSH = No preview
VNC/Monitor = Preview works
Enter fullscreen mode Exit fullscreen mode

Restart system 

sudo reboot
Enter fullscreen mode Exit fullscreen mode

🧠 Key Takeaways 

✔ Raspberry Pi OS now uses libcamera
✔ No manual camera enable needed
✔ rpicam-* commands replace raspistill
✔ SSH is for control, not display
✔ VNC gives full desktop access
Enter fullscreen mode Exit fullscreen mode

🎯 Final Thoughts

Once your Raspberry Pi and camera are set up:

👉 You are ready to:

  • Collect image datasets
  • Train AI models
  • Deploy computer vision applications

🔥 If you found this useful

  • Drop a ❤️
  • Share with someone learning Raspberry Pi
  • Follow for more AI + Embedded Systems content

Top comments (0)