Automated Timelapse System

A distributed timelapse video system using Raspberry Pi capture nodes and centralized video processing. Produces continuous daily timelapse videos from multiple camera angles with automatic golden hour detection.

Features

Multi-camera support – Currently configured for 2 IP cameras (Amcrest, Reolink)
Automatic video compilation – Real-time appending of video segments to daily files
Golden hour detection – Separate sunset videos using astronomical calculations
Staggered processing – Camera uploads separated by 30 seconds to prevent server overload
Period-based organization – Day (8am-4pm), Evening (4pm-12am), Night (12am-8am) periods
Location-aware – Sunset timing automatically adjusts for Douglas, Michigan coordinates
Resilient uploads – Retry logic with proper error handling

System Architecture

Raspberry Pi (Capture Node)

Image capture – RTSP stream capture every 15 seconds
Video processing – FFmpeg compilation of images into 15-minute segments
Upload client – cURL-based segment delivery to processing server
Staggered execution – Cameras process 30 seconds apart

Server (Processing Node)

Segment receiver – PHP-based upload handler
Video appending – FFmpeg concatenation of segments to daily videos
Period management – Automatic video rotation every 8 hours
Golden hour calculation – Astronomical sunset timing for special footage

Requirements

Raspberry Pi

Python 3.7+
FFmpeg 4.0+
RTSP-capable IP cameras
Reliable network connection

Server

PHP 7.4+ with file upload support
FFmpeg 4.0+
Web server (Apache/Nginx)
Sufficient storage (2-5GB per day)

Installation

Raspberry Pi Setup

# Clone repository
git clone https://github.com/username/timelapse-system.git
cd timelapse-system

# Create Python virtual environment
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

# Configure cameras and server
cp config.py.example config.py
# Edit config.py with your camera URLs and server details

Server Setup

# Copy PHP handler to web directory
cp upload.php /var/www/html/timelapse/

# Create required directories
mkdir -p /var/www/html/timelapse/videos
mkdir -p /var/www/html/timelapse/temp
chmod 755 /var/www/html/timelapse/videos
chmod 755 /var/www/html/timelapse/temp

# Ensure PHP can write to directories
chown -R www-data:www-data /var/www/html/timelapse/

Configuration

config.py (Raspberry Pi)

CAMERAS = {
    'amcrest': {
        'rtsp_url': 'rtsp://admin:password@192.168.1.100/stream',
        'name': 'amcrest',
        'stagger_offset': 0,  # Process at :00 seconds
        'enabled': True
    },
    'reolink': {
        'rtsp_url': 'rtsp://admin:password@192.168.1.101/stream', 
        'name': 'reolink',
        'stagger_offset': 30,  # Process at :30 seconds
        'enabled': True
    }
}

SERVER_CONFIG = {
    'upload_url': 'http://yourserver.com/timelapse/upload.php',
    'upload_retries': 3,
    'upload_timeout': 300
}

Geographic Configuration

Update coordinates in upload.php for accurate sunset calculation:

private $latitude = 42.5596;  // Your latitude
private $longitude = -86.2355; // Your longitude

Usage

Start Raspberry Pi System

# Run as service (recommended)
sudo systemctl enable timelapse
sudo systemctl start timelapse

# Or run manually
cd /path/to/timelapse
python3 main.py

Monitor System

# Check logs
tail -f /home/pi/timelapse/logs/timelapse_*.log

# View system status
curl http://yourserver.com/timelapse/upload.php?action=status

Manual Operations

# Force video rebuild
curl http://yourserver.com/timelapse/upload.php?action=rebuild&camera=amcrest&type=day

# Archive current videos  
curl http://yourserver.com/timelapse/upload.php?action=rotate

File Structure

timelapse-system/
├── main.py                 # Main controller
├── process_segments.py     # Video processing script
├── config.py              # Configuration
├── upload.php             # Server upload handler
├── README.md
├── requirements.txt
└── systemd/
    └── timelapse.service   # Systemd service file

API Endpoints

POST /upload.php

Upload video segment for processing

Parameters: camera, video file, timestamp, size
Response: JSON with status and video update info

GET /upload.php?action=status

Get system status and video information

GET /upload.php?action=rotate

Archive current videos and start fresh daily files

Daily Output

The system produces 8 video files per day:

{camera}_day_{YYYYMMDD}.mp4 – Daytime footage (8am-4pm)
{camera}_evening_{YYYYMMDD}.mp4 – Evening footage (4pm-12am)
{camera}_night_{YYYYMMDD}.mp4 – Overnight footage (12am-8am)
{camera}_sunset_{YYYYMMDD}.mp4 – Golden hour footage (sunset ±30min)

Videos update in real-time as segments arrive throughout the day.

Performance

Processing time: ~2-3 seconds per segment
Storage usage: ~2-5GB per camera per day
Network bandwidth: ~50KB per segment upload
Server load: Minimal (stream copy operations only)

Troubleshooting

Common Issues

No uploads reaching server

Check network connectivity and server URL
Verify camera RTSP streams are accessible
Review Raspberry Pi logs for processing errors

FFmpeg append failures

Ensure server has write permissions to video directories
Check FFmpeg installation and version compatibility
Review PHP error logs for specific FFmpeg error messages

Timing/stagger issues

Verify system clocks are synchronized (NTP)
Check configuration stagger_offset values
Monitor controller logs for trigger timing

Contributing

Fork the repository
Create a feature branch
Make your changes with appropriate tests
Submit a pull request with clear description

License

MIT License – See LICENSE file for details

Support

For issues and questions, please create a GitHub issue with:

System information (Pi model, server OS, versions)
Relevant log excerpts
Clear description of expected vs actual behavior

Timelapse,version 563,549