Add PiCam network streamer with auto-setup for Pi 5

Initial commit of the MJPEG streaming server for Raspberry Pi.
Supports Pi Camera Module (via picamera2), USB webcams (via OpenCV),
and auto-detection of camera type.

Includes:
- `stream_picamera2.py` — optimized Pi CSI camera streamer
- `stream_usb.py` — generic USB camera streamer
- `stream.py` — auto-detects camera and launches correct streamer
- `image-setup/` — SD card imaging scripts and Pi 5 headless setup
  guide for first-boot automation with systemd service
- `requirements.txt` — Python dependencies
This commit is contained in:
2026-05-23 10:39:31 +02:00
commit 6faf505f4b
10 changed files with 778 additions and 0 deletions
+81
View File
@@ -0,0 +1,81 @@
# PiCam Network Streamer
A lightweight MJPEG streaming server for Raspberry Pi. It serves a webpage you can open on any device on your local network to view the live camera feed.
## Pi 5 OS Image Setup
For a fully automated first-boot setup on a **Raspberry Pi 5** with an **IMX219** camera, see the [`image-setup/`](image-setup/) directory. It includes:
- `SETUP_GUIDE.md` — Step-by-step SD card imaging instructions.
- `autosetup.sh` — One-command setup script (installs drivers, dependencies, and auto-starts the streamer as a systemd service).
- `config_camera.txt``config.txt` snippet to force-enable the IMX219 overlay.
## Quick Start
1. **Install dependencies**
```bash
pip install -r requirements.txt
```
2. **Run the streamer**
*Auto-detect (recommended):*
```bash
python stream.py
```
*For the official Raspberry Pi Camera Module (CSI):*
```bash
python stream_picamera2.py
```
*For a USB webcam:*
```bash
python stream_usb.py
```
3. **View the stream**
Open a browser on any device on the same network and go to:
```
http://<raspberry-pi-ip>:5000
```
Replace `<raspberry-pi-ip>` with your Pi's IP address.
## Options
Both scripts accept the same command-line arguments:
| Argument | Description | Default |
|----------|-------------|---------|
| `--host` | Network interface to bind to | `0.0.0.0` |
| `--port` | Port to serve on | `5000` |
| `--width` | Video width in pixels | `640` |
| `--height` | Video height in pixels | `480` |
| `--fps` | Target framerate | `30` |
*USB streamer only:*
| `--device` | Video device index (e.g., `0` for `/dev/video0`) | `0` |
### Examples
Run at 720p on port 8080:
```bash
python stream_picamera2.py --width 1280 --height 720 --port 8080
```
Use a second USB camera:
```bash
python stream_usb.py --device 1
```
## Files
- `stream_picamera2.py` — Optimized streamer for Raspberry Pi Camera Module v2/v3 using `picamera2`.
- `stream_usb.py` — Generic streamer for USB cameras using OpenCV.
- `stream.py` — Auto-detects camera type and launches the correct streamer.
- `requirements.txt` — Python dependencies.
- `image-setup/` — Scripts and guides for preparing a headless Pi 5 OS image with IMX219 support.
- `SETUP_GUIDE.md` — Full imaging and setup instructions.
- `setup-pi.sh` — Run this on a live Pi to install everything and auto-start the streamer.
- `autosetup.sh` — For embedding into a custom SD card image (runs on first boot).
- `firstrun.sh` — Raspberry Pi first-boot trigger script.
- `config_camera.txt` — Snippet to enable the IMX219 overlay.
+121
View File
@@ -0,0 +1,121 @@
# Pi 5 + IMX219 OS Image Setup Guide
This guide walks you through setting up a **Raspberry Pi 5** with an **IMX219** camera module (Camera Module v2) to run the PiCam streamer.
## Option A: Easy Method — Run Setup on a Live Pi (Recommended)
If you already have Raspberry Pi OS running on your Pi 5:
1. **Copy the project** to your Pi (via USB drive, `scp`, or `git clone`).
2. **Run the setup script:**
```bash
cd piCam/image-setup
chmod +x setup-pi.sh
./setup-pi.sh
```
3. **Done.** The streamer will start automatically and will auto-start on every boot.
Open `http://<pi-ip>:5000` in your browser.
---
## Option B: Prepare an SD Card Image from Scratch
### 1. Download Raspberry Pi OS
Download the latest **Raspberry Pi OS (64-bit)** from the official site:
```
https://www.raspberrypi.com/software/operating-systems/
```
Recommended: **Raspberry Pi OS with Desktop** or **Lite** if you want a headless server.
### 2. Flash the OS to SD Card
Use the official **Raspberry Pi Imager**:
```
https://www.raspberrypi.com/software/
```
Steps in Imager:
1. Choose OS → Raspberry Pi OS (64-bit)
2. Choose Storage → Your SD card
3. Click the **gear icon (⚙️)** to open Advanced Options and configure:
- **Hostname**: `picam` (or your preference)
- **Enable SSH**: ☑️ (Use password authentication or public key)
- **Set username/password**: `pi` / your-password
- **Configure Wi-Fi**: Enter your SSID and password
- **Set locale**: Your timezone and keyboard layout
4. Click **Write**
### 3. Enable Camera Hardware
After flashing, the SD card will show a `bootfs` partition. You need to ensure camera support is enabled.
#### For Raspberry Pi OS Bookworm and later:
The IMX219 is usually auto-detected. However, if you need to force it, open `bootfs/config.txt` and add to the end:
```ini
# IMX219 Camera Module v2
dtoverlay=imx219
```
> **Note:** On Pi 5, the camera uses the CSI/DSI connector. Make sure the flat cable is oriented correctly (blue tape facing away from the USB ports on Pi 5).
### 4. Copy the Project Files
Copy the contents of this `image-setup` folder to the SD card:
```bash
# Linux/macOS example (adjust paths as needed)
cp -r /path/to/piCam/image-setup/* /media/YOUR_USERNAME/bootfs/
```
Specifically, ensure these files are placed on the `bootfs` partition:
- `autosetup.sh` → root of `bootfs/`
- `firstrun.sh` → root of `bootfs/`
### 5. Boot the Pi
Insert the SD card into the Pi 5, connect power, and wait 2-3 minutes for the first-boot setup to complete.
### 6. Access the Stream
Once booted, find the Pi on your network:
```bash
ping picam.local
```
Then open in a browser:
```
http://picam.local:5000
```
---
## Troubleshooting
| Issue | Solution |
|-------|----------|
| Camera not detected | Run `libcamera-hello --list-cameras` to verify hardware detection |
| Green tint on image | Normal under some lighting; adjust AWB in picamera2 config |
| Low framerate | Reduce resolution or ensure GPU memory isn't too low |
| `picamera2` not found | Run `sudo apt install -y python3-picamera2` |
## Automated First-Boot Setup
If you copied the provided files to `bootfs`, the system will:
1. Expand the filesystem
2. Update all packages
3. Install camera drivers, Python dependencies, and Flask
4. Enable the camera interface
5. Auto-start the streaming server on boot
Monitor progress by SSHing in:
```bash
ssh pi@picam.local
tail -f /var/log/picam-setup.log
```
+98
View File
@@ -0,0 +1,98 @@
#!/bin/bash
# PiCam First-Boot Setup Script
# Run this on the Raspberry Pi after first boot to configure the IMX219 camera
# and install the streaming server.
set -e
LOGFILE="/var/log/picam-setup.log"
PROJECT_DIR="/home/pi/piCam"
REPO_URL="" # Leave empty if copying files manually to the SD card
exec > >(tee -a "$LOGFILE")
exec 2>&1
echo "========================================"
echo "PiCam First-Boot Setup Starting"
echo "Date: $(date)"
echo "========================================"
# Update system
echo "[1/6] Updating package lists..."
sudo apt-get update
echo "[2/6] Installing system dependencies..."
sudo apt-get install -y \
python3-pip \
python3-venv \
python3-picamera2 \
libcamera-dev \
libcap-dev \
git \
i2c-tools \
v4l-utils
# Ensure camera interface is enabled (legacy method, though Bookworm uses libcamera)
echo "[3/6] Ensuring camera interface is enabled..."
if command -v raspi-config &> /dev/null; then
sudo raspi-config nonint do_camera 0 || true
fi
# Create project directory
echo "[4/6] Setting up project directory..."
mkdir -p "$PROJECT_DIR"
# If files are already present (e.g., copied to SD card), use them.
# Otherwise, clone or create a placeholder.
if [ ! -f "$PROJECT_DIR/stream_picamera2.py" ]; then
echo "Stream scripts not found in $PROJECT_DIR."
echo "Please copy the project files manually:"
echo " scp -r /path/to/piCam/* pi@picam.local:/home/pi/piCam/"
fi
# Setup Python virtual environment
echo "[5/6] Creating Python virtual environment..."
python3 -m venv "$PROJECT_DIR/venv"
source "$PROJECT_DIR/venv/bin/activate"
# Install Python dependencies
echo "[6/6] Installing Python packages..."
pip install --upgrade pip
pip install Flask opencv-python-headless
# Install systemd service for auto-start
echo "Installing systemd service..."
cat <<EOF | sudo tee /etc/systemd/system/picam-stream.service
[Unit]
Description=PiCam MJPEG Streamer
After=network-online.target
Wants=network-online.target
[Service]
Type=simple
User=pi
WorkingDirectory=$PROJECT_DIR
Environment="PATH=$PROJECT_DIR/venv/bin"
ExecStart=$PROJECT_DIR/venv/bin/python $PROJECT_DIR/stream_picamera2.py --host 0.0.0.0 --port 5000
Restart=always
RestartSec=5
[Install]
WantedBy=multi-user.target
EOF
sudo systemctl daemon-reload
sudo systemctl enable picam-stream.service
echo "========================================"
echo "Setup Complete!"
echo "========================================"
echo "The streaming server will start automatically on boot."
echo "To start it now, run: sudo systemctl start picam-stream"
echo "To check status: sudo systemctl status picam-stream"
echo "To view logs: sudo journalctl -u picam-stream -f"
echo ""
echo "Open in your browser: http://$(hostname -I | awk '{print $1}'):5000"
# Mark setup as complete so it doesn't run again
touch /home/pi/.picam-setup-done
+11
View File
@@ -0,0 +1,11 @@
# Add these lines to the end of your /boot/firmware/config.txt (or /boot/config.txt)
# to explicitly enable the IMX219 camera module on Raspberry Pi 5.
# IMX219 Camera Module v2
dtoverlay=imx219
# Ensure camera is enabled (legacy, usually not needed on Bookworm but safe)
camera_auto_detect=1
# Optional: Increase GPU memory if needed for higher resolutions
# gpu_mem=128
+20
View File
@@ -0,0 +1,20 @@
#!/bin/bash
# Raspberry Pi First-Boot Script
# Place this file on the bootfs partition as 'firstrun.sh' to execute on first boot.
# It will trigger the autosetup.sh script located in /boot/autosetup.sh
BOOT_SCRIPT="/boot/firmware/autosetup.sh"
# For older Raspberry Pi OS versions, the boot partition is mounted at /boot
if [ ! -f "$BOOT_SCRIPT" ]; then
BOOT_SCRIPT="/boot/autosetup.sh"
fi
if [ -f "$BOOT_SCRIPT" ]; then
echo "Found autosetup script at $BOOT_SCRIPT"
chmod +x "$BOOT_SCRIPT"
# Run as the default user (usually pi)
sudo -u pi bash "$BOOT_SCRIPT"
else
echo "No autosetup.sh found on boot partition."
fi
+110
View File
@@ -0,0 +1,110 @@
#!/bin/bash
# PiCam Live Setup — Run this directly on your Raspberry Pi 5 after first boot.
# This automates installation of dependencies and enables auto-start streaming.
set -e
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
PROJECT_DIR="$HOME/piCam"
LOGFILE="/var/log/picam-setup.log"
mkdir -p "$(dirname "$LOGFILE")"
exec > >(tee -a "$LOGFILE")
exec 2>&1
echo "============================================"
echo " PiCam Setup Script"
echo " Running on: $(hostname)"
echo " Date: $(date)"
echo "============================================"
# 1. System update
echo "[1/7] Updating package lists..."
sudo apt-get update
# 2. Install system packages
echo "[2/7] Installing system packages..."
sudo apt-get install -y \
python3-pip \
python3-venv \
python3-picamera2 \
libcamera-dev \
libcap-dev \
git \
i2c-tools \
v4l-utils
# 3. Detect camera
echo "[3/7] Detecting camera..."
if command -v libcamera-hello >/dev/null 2>&1; then
echo "libcamera devices:"
libcamera-hello --list-cameras || true
else
echo "libcamera-hello not found; relying on picamera2 detection."
fi
# 4. Copy project files
echo "[4/7] Installing project files..."
mkdir -p "$PROJECT_DIR"
if [ -d "$SCRIPT_DIR/../" ] && [ -f "$SCRIPT_DIR/../stream_picamera2.py" ]; then
# Running from cloned repo
cp "$SCRIPT_DIR/../stream_picamera2.py" "$PROJECT_DIR/"
cp "$SCRIPT_DIR/../stream_usb.py" "$PROJECT_DIR/"
cp "$SCRIPT_DIR/../stream.py" "$PROJECT_DIR/"
cp "$SCRIPT_DIR/../requirements.txt" "$PROJECT_DIR/"
else
echo "WARNING: Could not find project Python files."
echo "Please copy them manually to $PROJECT_DIR"
fi
# 5. Python environment
echo "[5/7] Creating Python virtual environment..."
python3 -m venv "$PROJECT_DIR/venv"
source "$PROJECT_DIR/venv/bin/activate"
pip install --upgrade pip
pip install Flask opencv-python-headless
# 6. Create systemd service
echo "[6/7] Creating systemd service..."
cat <<EOF | sudo tee /etc/systemd/system/picam-stream.service
[Unit]
Description=PiCam MJPEG Streamer
After=network-online.target
Wants=network-online.target
[Service]
Type=simple
User=$USER
WorkingDirectory=$PROJECT_DIR
Environment="PATH=$PROJECT_DIR/venv/bin"
ExecStart=$PROJECT_DIR/venv/bin/python $PROJECT_DIR/stream_picamera2.py --host 0.0.0.0 --port 5000
Restart=always
RestartSec=5
[Install]
WantedBy=multi-user.target
EOF
sudo systemctl daemon-reload
sudo systemctl enable picam-stream.service
# 7. Start service
echo "[7/7] Starting streamer..."
sudo systemctl start picam-stream.service
echo "============================================"
echo " Setup Complete!"
echo "============================================"
echo ""
echo "Streamer is running as a systemd service."
echo ""
echo "Commands:"
echo " Status: sudo systemctl status picam-stream"
echo " Logs: sudo journalctl -u picam-stream -f"
echo " Stop: sudo systemctl stop picam-stream"
echo " Start: sudo systemctl start picam-stream"
echo ""
IP=$(hostname -I | awk '{print $1}')
echo "Open in browser: http://$IP:5000"
+3
View File
@@ -0,0 +1,3 @@
Flask>=2.0.0
opencv-python-headless>=4.5.0
picamera2>=0.3.9; platform_machine=='aarch64' or platform_machine=='armv7l'
+32
View File
@@ -0,0 +1,32 @@
#!/usr/bin/env python3
"""Auto-detect and start the best available streamer."""
import importlib.util
import subprocess
import sys
def main():
# Check if picamera2 is available and a Pi camera is connected
if importlib.util.find_spec("picamera2") is not None:
try:
from picamera2 import Picamera2
# Attempt to create a camera to see if hardware is present
cam = Picamera2()
cam.close()
print("Detected Raspberry Pi camera. Starting picamera2 streamer...")
subprocess.run([sys.executable, "stream_picamera2.py"] + sys.argv[1:])
return
except Exception as e:
print(f"picamera2 available but camera not detected ({e}).")
else:
print("picamera2 not installed.")
# Fall back to USB / OpenCV
print("Falling back to USB camera streamer...")
subprocess.run([sys.executable, "stream_usb.py"] + sys.argv[1:])
if __name__ == "__main__":
main()
+141
View File
@@ -0,0 +1,141 @@
from flask import Flask, Response, render_template_string
from picamera2 import Picamera2
import io
import logging
import threading
import time
app = Flask(__name__)
# Configure logging
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
class Camera:
def __init__(self, resolution=(640, 480), framerate=30):
self.resolution = resolution
self.framerate = framerate
self.picam2 = None
self.frame = None
self.lock = threading.Lock()
self.running = False
self.thread = None
def start(self):
self.picam2 = Picamera2()
config = self.picam2.create_video_configuration(
main={"size": self.resolution, "format": "RGB888"},
controls={"FrameRate": self.framerate}
)
self.picam2.configure(config)
self.picam2.start()
# Allow camera to warm up
time.sleep(2)
self.running = True
self.thread = threading.Thread(target=self._capture, daemon=True)
self.thread.start()
logging.info("Camera started.")
def _capture(self):
while self.running:
try:
arr = self.picam2.capture_array()
# Encode as JPEG
import cv2
ret, buf = cv2.imencode('.jpg', arr)
if ret:
with self.lock:
self.frame = buf.tobytes()
time.sleep(0.001) # Brief sleep to yield
except Exception as e:
logging.error(f"Capture error: {e}")
time.sleep(0.1)
def get_frame(self):
with self.lock:
return self.frame
def stop(self):
self.running = False
if self.thread:
self.thread.join(timeout=2.0)
if self.picam2:
self.picam2.stop()
self.picam2.close()
logging.info("Camera stopped.")
camera = None
HTML_PAGE = """
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>PiCam Stream</title>
<style>
body {
background: #111;
color: #eee;
font-family: system-ui, -apple-system, sans-serif;
display: flex;
flex-direction: column;
align-items: center;
justify-content: center;
min-height: 100vh;
margin: 0;
}
h1 { margin-bottom: 0.5rem; }
p { color: #888; margin-bottom: 1rem; }
img {
max-width: 95vw;
max-height: 80vh;
border: 2px solid #333;
border-radius: 8px;
background: #000;
}
</style>
</head>
<body>
<h1>PiCam Live Stream</h1>
<p>{{ resolution }} @ {{ framerate }} fps</p>
<img src="{{ url_for('video_feed') }}" alt="Live stream">
</body>
</html>
"""
@app.route('/')
def index():
return render_template_string(HTML_PAGE, resolution=camera.resolution, framerate=camera.framerate)
@app.route('/video_feed')
def video_feed():
def generate():
while True:
frame = camera.get_frame()
if frame is None:
time.sleep(0.01)
continue
yield (b'--frame\r\n'
b'Content-Type: image/jpeg\r\n\r\n' + frame + b'\r\n')
return Response(generate(), mimetype='multipart/x-mixed-replace; boundary=frame')
if __name__ == '__main__':
import argparse
parser = argparse.ArgumentParser(description='PiCam MJPEG Streamer using picamera2')
parser.add_argument('--host', default='0.0.0.0', help='Interface to bind (default: 0.0.0.0)')
parser.add_argument('--port', type=int, default=5000, help='Port to bind (default: 5000)')
parser.add_argument('--width', type=int, default=640, help='Capture width')
parser.add_argument('--height', type=int, default=480, help='Capture height')
parser.add_argument('--fps', type=int, default=30, help='Target framerate')
args = parser.parse_args()
camera = Camera(resolution=(args.width, args.height), framerate=args.fps)
camera.start()
try:
logging.info(f"Starting server on http://{args.host}:{args.port}")
app.run(host=args.host, port=args.port, threaded=True, debug=False)
except KeyboardInterrupt:
pass
finally:
camera.stop()
+161
View File
@@ -0,0 +1,161 @@
import logging
import threading
import time
import cv2
from flask import Flask, Response, render_template_string
app = Flask(__name__)
logging.basicConfig(
level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s"
)
class UsbCamera:
def __init__(self, device_index=0, resolution=(640, 480), framerate=30):
self.device_index = device_index
self.resolution = resolution
self.framerate = framerate
self.cap = None
self.frame = None
self.lock = threading.Lock()
self.running = False
self.thread = None
def start(self):
self.cap = cv2.VideoCapture(self.device_index)
if not self.cap.isOpened():
raise RuntimeError(f"Cannot open camera device {self.device_index}")
self.cap.set(cv2.CAP_PROP_FRAME_WIDTH, self.resolution[0])
self.cap.set(cv2.CAP_PROP_FRAME_HEIGHT, self.resolution[1])
self.cap.set(cv2.CAP_PROP_FPS, self.framerate)
# Give the driver a moment to apply settings
time.sleep(0.5)
self.running = True
self.thread = threading.Thread(target=self._capture, daemon=True)
self.thread.start()
logging.info("USB camera started.")
def _capture(self):
while self.running:
ret, img = self.cap.read()
if not ret:
logging.warning("Failed to read frame from camera.")
time.sleep(0.05)
continue
ret2, buf = cv2.imencode(".jpg", img)
if ret2:
with self.lock:
self.frame = buf.tobytes()
# Throttle slightly to avoid maxing CPU
time.sleep(1.0 / max(self.framerate, 1))
def get_frame(self):
with self.lock:
return self.frame
def stop(self):
self.running = False
if self.thread:
self.thread.join(timeout=2.0)
if self.cap:
self.cap.release()
logging.info("USB camera stopped.")
camera = None
HTML_PAGE = """
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>PiCam Stream</title>
<style>
body {
background: #111;
color: #eee;
font-family: system-ui, -apple-system, sans-serif;
display: flex;
flex-direction: column;
align-items: center;
justify-content: center;
min-height: 100vh;
margin: 0;
}
h1 { margin-bottom: 0.5rem; }
p { color: #888; margin-bottom: 1rem; }
img {
max-width: 95vw;
max-height: 80vh;
border: 2px solid #333;
border-radius: 8px;
background: #000;
}
</style>
</head>
<body>
<h1>PiCam Live Stream</h1>
<p>{{ resolution }} @ {{ framerate }} fps</p>
<img src="{{ url_for('video_feed') }}" alt="Live stream">
</body>
</html>
"""
@app.route("/")
def index():
return render_template_string(
HTML_PAGE, resolution=camera.resolution, framerate=camera.framerate
)
@app.route("/video_feed")
def video_feed():
def generate():
while True:
frame = camera.get_frame()
if frame is None:
time.sleep(0.01)
continue
yield (b"--frame\r\nContent-Type: image/jpeg\r\n\r\n" + frame + b"\r\n")
return Response(generate(), mimetype="multipart/x-mixed-replace; boundary=frame")
if __name__ == "__main__":
import argparse
parser = argparse.ArgumentParser(
description="PiCam MJPEG Streamer using OpenCV (USB camera)"
)
parser.add_argument(
"--host", default="0.0.0.0", help="Interface to bind (default: 0.0.0.0)"
)
parser.add_argument(
"--port", type=int, default=5000, help="Port to bind (default: 5000)"
)
parser.add_argument(
"--device", type=int, default=0, help="Video device index (default: 0)"
)
parser.add_argument("--width", type=int, default=640, help="Capture width")
parser.add_argument("--height", type=int, default=480, help="Capture height")
parser.add_argument("--fps", type=int, default=30, help="Target framerate")
args = parser.parse_args()
camera = UsbCamera(
device_index=args.device,
resolution=(args.width, args.height),
framerate=args.fps,
)
camera.start()
try:
logging.info(f"Starting server on http://{args.host}:{args.port}")
app.run(host=args.host, port=args.port, threaded=True, debug=False)
except KeyboardInterrupt:
pass
finally:
camera.stop()