ALPR Parking Time Service¶
The ALPR Parking Time service is designed for mobile camera systems mounted inside moving vehicles. It enables parking zone monitoring, parking time enforcement, and vehicle presence tracking by continuously scanning and recognizing license plates while driving through designated areas.
Use Cases¶
- Parking zone enforcement - Monitor vehicle parking duration in paid zones
- Parking lot management - Track vehicle entry/exit and dwell times
- Street parking audits - Survey parked vehicles along routes
- Permit verification - Identify vehicles with/without valid parking permits
- Time-limited parking monitoring - Detect overstays in restricted areas
Camera Requirements¶
In-Vehicle Mounted Cameras¶
Cameras installed behind the windshield or on vehicle roof, scanning parked vehicles while driving.
Video requirements: - Frame rate: 50-60 FPS (higher frame rate is essential) - Resolution: Full HD (1920×1080) minimum - Vehicle speed: < 30 km/h for reliable recognition - Wide-angle lens recommended for side-parked vehicles
Why high frame rate? - Moving camera + stationary plates = motion blur risk - Higher FPS provides more frames per vehicle - Increases chance of capturing sharp, readable plate image
Mounting considerations: - Stable mounting to minimize vibration - Anti-glare filters for windshield mounting - Multiple cameras for left/right side coverage
Generated Reports and Data¶
The parking time service provides specialized reports for parking analysis.
Output Files¶
| File | Description |
|---|---|
History.csv |
Complete scan history with timestamps |
Alpr.csv |
All license plate readings with metadata |
AlprOccurences.csv |
Plate occurrence matrix by time intervals |
AlprParktime.csv |
Parking duration analysis per vehicle |
Alpr0to500.xlsx, ... |
Excel sheets with images |
Parking Time Analysis¶
The AlprParktime.csv file contains calculated parking durations:
| Field | Description |
|---|---|
| License Plate | Recognized plate text |
| Start | First detection timestamp |
| End | Last detection timestamp |
| Duration | Calculated parking time (minutes) |
This data is grouped by default 5-minute intervals and sorted by duration (longest first).
Data Fields¶
Each license plate detection includes:
| Field | Description |
|---|---|
ALPR |
Recognized license plate text |
ALPR_score |
Recognition confidence (0.0 - 1.0) |
ALPR_type |
License plate type classification |
ALPR_region |
Region/country code |
ALPR_lettercode |
District/city letter code |
Czas (Time) |
Detection timestamp |
Image |
Vehicle image path |
ALPR_image |
Cropped plate image path |
Occurrence Matrix¶
The AlprOccurences.csv provides a time-based matrix showing when each license plate was detected:
| Time | ABC1234 | XYZ5678 | ... |
|---|---|---|---|
| 08:00 | 1 | 0 | ... |
| 08:05 | 1 | 1 | ... |
| 08:10 | 0 | 1 | ... |
This matrix helps visualize vehicle presence patterns across time.
Workflow Example¶
- Route planning - Define patrol route through parking zone
- Data collection - Drive route with camera recording (repeat every X minutes)
- Processing - Upload videos to AITracker for ALPR analysis
- Parking time calculation - System correlates same plates across multiple passes
- Enforcement report - Export vehicles exceeding time limits
Accuracy¶
| Factor | Impact on Accuracy |
|---|---|
| Vehicle speed < 30 km/h | Optimal |
| Vehicle speed 30-50 km/h | Reduced accuracy |
| Night-time operation | Requires IR camera |
| Angle to parked vehicles | 15-45° optimal |
Overall ALPR accuracy: ~90% under optimal conditions
Processing speed: x1-x2 real-time for 50-60 FPS video
Configuration¶
Configure the parking time service:
[Service]
id = "ServiceAlprParktime"
[ALPR]
enabled = true
save_images = true
grouping_period = "5min" # Time grouping for occurrence analysis
Limitations¶
- GPS integration - Currently timestamps only; GPS correlation requires external tools
- Multiple passes - Parking time calculation requires multiple route passes
- Plate visibility - Obstructed or dirty plates may not be recognized