Skip to main content

Dahua Integration

The integration with Dahua devices enables the use of facial recognition and LPR (license plate recognition) equipment in ACCELERO access control.

Licensing Requirement

This feature requires the Dahua module enabled in the license. See Licensing.

Overview​

The Dahua integration supports two main features:

FeatureDescription
Facial RecognitionAutomatic synchronization of facial templates to Dahua facial identification devices
LPR (License Plate Recognition)Automatic capture of vehicle license plates with confidence level control
Use without physical controllersPossibility of using Dahua/Intelbras controllers directly, without the need for additional physical controllers
Architecture

The integration is done through an ACCELERO/DAHUA communication gateway (IONGRADE product). The gateway bridges the ACCELERO servers and the Dahua devices.

Navigation path: Advanced > System > Integrations - Dahua

Homologated models​

TypeModelIntegration modeNotes
FacialDHI-ASI3213ANeon ControllerDirect planned (coming soon); check the Wiegand output
LPRITC215-PW6M-xxx, ITC215-PW4I-xxxNeon Controller—

Explanation of the modes (Direct × Neon Controller) and the complete matrix in Homologated Equipment.

Facial Recognition​

Architecture and Requirements​

The integration with Dahua facial identification devices uses:

  1. ACCELERO/DAHUA Gateway (IONGRADE product)
  2. Dahua Devices for facial recognition
  3. ACCELERO Server with the module enabled

Data flow:

ACCELERO → Gateway → Dahua Devices
        ← Status ←

General Settings​

Path: Advanced > System > Integrations - Dahua > Settings

Main Parameters​

FieldDescription
TypeDahua device model
Categories that must have facial recognitionDefines which categories will have facial templates generated and distributed.
Example: disable for visitors
Automatically remove templates for the following categoriesCategories whose templates are automatically deleted after passing through controllers with "Automatic visitor removal"
Except if the person has at least one of the following categoriesException categories - people with these categories will never have their templates removed automatically
Device Memory

Automatic template removal is essential to avoid fully occupying the memory of the Dahua devices, especially in systems with high visitor traffic.

TODO: Add Screenshot

Include a screenshot of the general settings screen showing all the fields described above.

Registering Gateways​

Path: Settings > Dahua > Gateways

For each communication gateway, configure:

FieldDescription
Serial numberSerial number of the gateway (obtain from the physical equipment).
CRITICAL: The information must be correct or communication will fail
ModelModel of the communication gateway
DescriptionName/description of the gateway

Monitoring:

  • ACCELERO ↔ Gateway communication status
  • Gateway ↔ subordinate devices communication status
TODO: Add Screenshot

Include a screenshot of the gateway registration screen showing fields and communication status.

Registering Devices​

Path: Settings > Dahua > Devices

For each Dahua facial recognition device:

FieldDescription
DescriptionModel/location of the device
IP addressIP of the device on the network
Communication portTCP port for communication
UserUser for authentication on the device
PasswordAuthentication password
GatewayCommunication gateway to which the device is subordinate
Synchronize only these categoriesAdditional filter of categories (in addition to the global filter).
If empty, it synchronizes all categories configured globally
ActiveIf disabled, the device will not receive registration/deletion commands
Double Category Filter

The category filter works on two levels:

  1. Global (Settings): Defines allowed categories in the system
  2. Per device: Further filters which categories go to that specific device

Only categories that pass both filters will be synchronized.

TODO: Add Screenshot

Include a screenshot of the device registration screen showing all fields.

How Facial Synchronization Works​

Automatic Synchronization​

After the settings, synchronization happens automatically:

  1. When saving a person's registration:

    • The system checks whether the person has a category configured for facial recognition
    • If so, it generates a template from the first photo of the registration
    • It distributes the template to all active devices

  2. Visual indicators on the photo:

    • 🔴 Red icon: Templates not yet fully distributed
    • 🔵🔵 Two blue icons: Templates distributed successfully
Photo Quality

The first photo of the registration is always used for synchronization. Make sure:

  • ✓ Good capture quality
  • ✓ Good lighting
  • ✓ Frontal framing of the person
  • ✓ Neutral background (if possible)

Manual Synchronization (Resend)​

Path: Advanced > System > Integrations - Dahua > Synchronize templates

When to use:

  • Replacing a device in the field
  • Restoring after a failure
  • Bulk update

How to do it:

  1. Select the desired devices
  2. Click Resend templates
  3. Wait for completion (it can take considerable time)
Processing Time

The resend time depends on the number of photos. For large systems (thousands of people), the process can take hours.

TODO: Add Screenshot

Include a screenshot of the template synchronization screen showing device selection and the resend button.

Automatic Visitor Removal​

ACCELERO can automatically delete visitor templates after their exit to avoid fully occupying the device memory.

How It Works​

When the deletion occurs:

  1. When passing through a barrier with the "Remove visitors" flag:

    • The system checks whether the person has a category from the "Automatically remove" list
    • It checks whether the person does NOT have a category from the "Except if..." list
    • If the conditions are met, it deletes the template from the devices

  2. When closing/deleting an event:

    • The same check above is performed
    • Templates are removed automatically

Configuration example:

Automatically remove: VISITORS, CONTRACTORS
Except if the person has: EMPLOYEES, RESIDENTS

Result:
- Pure visitor: template will be deleted after exit ✓
- Visitor who is also an employee: template kept ✗
Memory Management

Automatic removal is essential in systems with high visitor traffic to:

  • Avoid exhausting device memory
  • Keep only active registrations
  • Ensure facial recognition performance

LPR (Vehicle License Plate Recognition)​

Architecture and Requirements​

The Dahua LPR integration also uses the ACCELERO/DAHUA gateway.

Data flow:

LPR Device → Gateway → ACCELERO
                     → Generates access event
How It Works

When the LPR identifies a plate, the system generates an access request on the configured controller/channel, applying the same access policy rules.

Gateway Configuration​

Path: Settings > Dahua > Gateways

Same configuration as facial recognition (shared gateway).

Registering LPR Devices​

Path: Settings > Dahua > LPR Devices

For each Dahua LPR device:

FieldDescription
DescriptionModel/location of the LPR device
Minimum confidence levelQuality filter: minimum level (0 to 100) for a plate reading to be considered valid.
Typical values: 70-85
Serial number/identifierSerial number or identifier configured on the LPR device
Associated controllerController/channel where identified plates will be reported as access attempts
GatewayCommunication gateway to which the device is subordinate
Confidence Level

The "Minimum confidence level" field is exclusive to the Dahua integration and allows filtering out low-quality readings.

Recommendations:

  • Controlled environments (good lighting, clean plates): 80-90
  • Normal environments: 70-80
  • Challenging environments (rain, low light): 60-70

Values that are too low can generate false positives. Values that are too high can reject valid readings.

Associated Controller

The "Associated controller" field is critical: it defines which system channel will receive the access events when a plate is identified.

The access policy configuration of this channel will determine whether the vehicle is granted or denied access.

TODO: Add Screenshot

Include a screenshot of the LPR device registration screen showing the confidence level field.

How the LPR Integration Works​

Operation Flow​

  1. Dahua LPR identifies a vehicle plate
  2. Checks the confidence level:
    • If confidence < configured minimum: discards the reading ✗
    • If confidence ≥ configured minimum: proceeds ✓
  3. The gateway notifies ACCELERO
  4. ACCELERO generates an access request on the configured controller/channel
  5. The system applies the access rules:
    • Checks whether the plate is registered
    • Checks whether the vehicle is enabled
    • Checks the access policies (category × area × schedule)
    • Applies validity rules
  6. Result:
    • ✅ Access authorized: Opening command sent
    • ❌ Access denied: Event recorded, no opening

Visualization:

  • The Event Log shows all the information normally
  • The identified plate and confidence level appear in the log
  • The authorization result is recorded

See Vehicles for registering vehicles and plates.

Equivalence to a Card

The plate identification by the LPR is treated by the system as equivalent to a card passage on the configured channel. All access policy rules are applied normally.

Best Practices​

Planning​

  1. Define categories before configuring:

    • Which categories will have facial recognition?
    • Which categories will have automatic removal?
    • Which categories will be exceptions?

  2. Size the device memory:

    • Check the template capacity per device
    • Plan the automatic removal policy
    • Monitor memory usage

  3. Configure the LPR confidence level:

    • Test different values in your environment
    • Adjust according to the false positive/negative rate
    • Document the chosen value

Operation​

  1. Photo quality:

    • Train operators on the importance of the first photo
    • Use adequate lighting
    • Always frontal framing

  2. Monitor synchronization:

    • Check the status icons on the photos
    • Investigate synchronization failures quickly
    • Use resend only when necessary

  3. Manage memory:

    • Enable automatic removal for visitors
    • Configure exception categories correctly
    • Monitor device memory usage

  4. LPR - Fine-tuning:

    • Analyze the logs to identify rejected readings
    • Adjust the confidence level as needed
    • Keep plates updated in the system
    • Configure appropriate access policies

Maintenance​

  1. Check communication status:

    • Gateway ↔ ACCELERO
    • Gateway ↔ Devices
    • Resolve network problems quickly

  2. Back up before changes:

    • Back up before reconfiguring gateways
    • Document configuration changes
    • Test after changes

  3. Device replacement:

    • Use the template resend function
    • Verify complete synchronization
    • Test recognition after replacement

  4. Firmware updates:

    • Consult IONGRADE before updating Dahua firmware
    • Check compatibility
    • Test in a controlled environment first

Troubleshooting​

Templates do not synchronize​

Possible causes:

  • Gateway without communication with ACCELERO
  • Gateway without communication with the device
  • Device marked as inactive
  • Person without a category configured for facial recognition
  • Network/credential problems

Solution:

  1. Check the communication status on the gateways screen
  2. Confirm that the device is active
  3. Check whether the person has the correct category
  4. Test the access credentials to the device
  5. Use manual resend if necessary

LPR rejects many valid readings​

Possible cause:

  • Confidence level too high for the environment conditions

Solution:

  1. Consult the event log to see the rejected readings
  2. Gradually reduce the confidence level (e.g., 80 → 75 → 70)
  3. Test and monitor the accuracy rate
  4. Find a balance between sensitivity and accuracy

LPR accepts incorrect readings​

Possible cause:

  • Confidence level too low

Solution:

  1. Gradually increase the confidence level
  2. Improve the lighting of the capture area
  3. Clean the camera lenses
  4. Check the capture angle

Device memory full​

Solution:

  1. Enable "Automatic template removal"
  2. Configure the correct categories
  3. Perform manual cleanup via the Dahua interface
  4. Consider increasing memory capacity (hardware)

Comparison: Dahua vs Hikvision​

Similarities​

  • Both use an IONGRADE gateway
  • Automatic template synchronization
  • Automatic visitor removal
  • Support for LPR and facial recognition

Main Differences​

FeatureHikvisionDahua
Virtual card formatConfigurable (including Wiegand 34-bit)Not available
Daily orphan cleanupConfigurableNot mentioned
LPR - Confidence levelNot available✓ Configurable (0-100)
Photo capture✓ AvailableNot mentioned
Choosing the Manufacturer

The choice between Hikvision and Dahua should consider:

  • Equipment already installed
  • Need for fine LPR control (favors Dahua)
  • Need for integrated photo capture (favors Hikvision)
  • Local manufacturer support

Next Steps​

  • Vehicles - Register vehicles for use with LPR
  • Categories - Configure categories for facial recognition
  • Controllers - Configure channels for LPR
  • People - Register photos for facial recognition
  • Licensing - Check available modules
  • Hikvision - Compare with the Hikvision integration