Key Capabilities:
People Counting:
- Two-line crossing counter with entry/exit detection using verify and confirm lines
- Anti-jitter logic with configurable deadzone, cooldown and deduplication
- Per-event snapshot capture for verification of every entry and exit
- Multi-object tracking maintains unique IDs per person across frames
- 15+ frames per second processing speed with 95%+ detection accuracy
- Configurable confirm frames for entry and exit events separately
Driver Monitoring:
- Drowsiness detection via advanced facial analysis for real-time drowsiness alerts
- Phone usage detection using dedicated AI detection
- Smoking detection with a dedicated AI model
- Instant real-time alerts with cooldown logic to prevent duplicate notifications
- Automatic alert snapshot capture for evidence logging
- 10+ FPS processing speed with 95%+ detection accuracy
Bus Session Management:
- Link driver and conductor employees to bus sessions with unique employee IDs
- All counting and alert events automatically tied to the active session
- Shift scheduling with configurable start/stop times and break periods
- Automated camera start/stop based on shift schedules
- Employee directory with name, phone and position (Driver/Conductor)
- Session status tracking (Active/Closed) with creation timestamps
Reports & Alerts:
- PDF, DOCX or CSV reports for people counting, driver alerts, or combined data
- Custom date range selection for any reporting period
- Driver alert snapshot gallery filtered by date and alert type
- Hourly and daily distribution charts for counting and driver events
- Summary statistics with entry/exit totals and alert breakdowns
- Scheduled and on-demand report generation
Video Recording & Playback:
- Record both people counting and driver monitoring camera feeds to disk
- Browse and playback recordings organised by day and hour
- Continuous recording with configurable segment duration
- Video Center page for easy navigation and review
- Delete individual recordings for storage management
- Decoupled recording pipeline — never impacts live inference performance
Live Dashboard:
- Real-time MJPEG video feed for both counting and driver cameras
- Live counters for entry, exit and total with instant WebSocket updates
- Interactive charts showing hourly activity, in vs out, and alert trends
- Recent event timeline with snapshot previews
- FPS, uptime and system status indicators
- Accessible from any device on the network via web browser
People Counting — How It Works:
- Capture — Video frames captured from RTSP camera or video file at configured frame rate.
- Detect — Hardware-accelerated AI model detects all persons within the Region of Interest.
- Track — Tracker assigns and maintains a unique ID for each person across consecutive frames.
- Count — When a person crosses the counting line, direction analysis determines entry or exit with anti-jitter protections.
- Record & Report — Events stored in database, snapshots saved, video recorded. Generate PDF, Word, or CSV reports.
Driver Monitoring — How It Works:
- Capture — Driver-facing camera captures video frames continuously.
- Analyze — AI models analyze facial features for drowsiness and detect phone and cigarette objects.
- Alert — Confirmed detections trigger real-time alerts for drowsy, phone usage, or smoking with cooldown to prevent duplicates.
- Record & Report — Alert snapshots saved, events logged to database. Generate PDF, Word, or CSV reports for any time period.
Feature Details:
Detection & Tracking:
- Advanced AI detection model optimised for edge hardware acceleration
- Dedicated AI model for smoking detection
- Advanced facial landmark analysis for drowsiness detection
- Multi-object tracker with persistent person IDs across frames
- Multi-core AI acceleration with parallel processing for both pipelines
- Dynamic core allocation between people counting and driver monitoring pipelines
Platform & Infrastructure:
- Hardware: compact, fanless edge AI device
- OS: Linux-based embedded operating system
- Database: secure relational database for events, sessions and configuration
- Streaming: live video and event streaming over standard web protocols
- Frontend: responsive web dashboard with interactive analytics charts
Counting Engine:
- Two-line crossing counter with verify (cyan) and confirm (red) lines
- Configurable deadzone, cooldown and deduplication parameters for accuracy tuning
- Multi-frame confirmation for reliable entry and exit detection
- Configurable inter-line crossing timeout to prevent stale tracks
- ROI filtering and frame skip for performance tuning
- Direction swap support for flexible camera installation angles
Licensing & Security:
- Cryptographically signed license files bound to device hardware ID
- Unique device fingerprint prevents unauthorised license copying
- Per-feature gating — individual modules licensed independently
- License status dashboard with colour-coded expiry warnings
- Role-based access — Admin (full access) and User (read-only with report generation)
- Session-based authentication with REST API for all operations
Deployment & Licensing:
Installation & Setup:
- Automated setup installs all dependencies and configures the system
- Provisions the database and runtime environment automatically
- Installs as a managed service for auto-start on boot
- Pre-configured RK3588 edge AI runtime environment
- Supports RTSP IP cameras and video files for testing
- Full setup from unboxing to first detection in under an hour
Web Interface (18 Pages):
- Dashboard with live dual-camera feeds, counters, charts and event timeline
- Session management — link drivers and conductors to active bus sessions
- Camera config with ROI drawing on live preview canvas
- Configuration for detection thresholds, counting lines and overlay toggles
- Employee management, accounts, shift scheduling and driver alert gallery
- Reports, recordings, logs, network config, system info and license management
What You Need / What We Provide:
- Hardware: compact edge AI device
- Camera: RTSP-compatible IP camera for counting and driver-facing camera
- Network: local Ethernet or Wi-Fi connection
- Browser: any modern web browser (Chrome, Firefox, Edge, Safari)
- We provide: pre-configured edge device, pre-installed TrackSense-AI software, installation assistance, calibration and operator training, software updates during license period
API & Integration:
- Full REST API covering authentication, counting, driver monitoring, sessions, reports, recordings, charts and system management
- WebSocket for real-time event notifications and live video frames
- Comprehensive API endpoints across all functional categories
- Secure session authentication for browser and client access
- MJPEG video feed endpoint for embedding in external monitoring tools
- Health and node status endpoints for external system polling