Best Practices for AI

Topics included in this article:

• Overview
• Initial AI Expectations
• Camera Setup for Optimal AI Accuracy
  • Camera Positioning 
• Placing Activity Regions
• Human Movement
• Facial Recognition
• Vehicle Movement
• License Plate Recognition
• Helpful Links
• Contact Support or Sales

Overview

Rhombus cameras use artificial intelligence to automatically detect and track objects such as people, faces, vehicles, and license plates. These AI-powered features help users identify important activity, search events, and investigate footage through the Rhombus Console.

Note: Relevant event types must be enabled and included in the Alert Policy assigned to the camera to generate alerts for detected events.

Initial AI Expectations

Since camera environments and applications vary, AI thresholds may need to be adjusted to fine-tune event capture. For more information, see the AI Confidence Settings article.

Camera Setup for Optimal AI Accuracy

Some AI features have additional setup requirements. For more information, see the article for the feature you are configuring:

• Managing Facial Recognition
• Managing License Plate Recognition (LPR)

Camera Positioning

Lighting, camera angle, distance, and positioning can impact the accuracy of AI event detection.

• When mounting a camera, minimize unused space in the camera's field of view.
• Ensure the camera is mounted at an appropriate distance for the camera model and AI detection type. For more information, see Optics and Object Distances for Analytics.
• Ensure there is sufficient lighting and avoid backlighting, particularly when using features such as facial recognition and license plate recognition.

Placing Activity Regions

For camera views with high levels of activity, such as public spaces, activity regions can help reduce the number of events generated.

Activity regions can be configured in two ways:

• Define an area of interest where you want to capture events, such as a building entrance.
• Define an area where you want to exclude event capture, such as an area with predominantly foliage or a public road.

For more information, see the Region Setup article.

Human Movement

Facial Recognition

Facial recognition occurs after the system detects human movement. If the camera is not positioned to reliably detect human movement, facial recognition may also be impacted. Additional camera placement requirements can help optimize facial recognition performance.

For more information, see the Managing Facial Recognition article.

Vehicle Movement

For optimal vehicle movement detection, position the camera so the entire vehicle enters the field of view. If only a portion of the vehicle is visible, vehicle movement detection may not trigger.

Vehicle speed can also impact detection accuracy. Ensure the camera has sufficient time to capture the vehicle as it moves through the field of view so the AI can process and identify the vehicle.

License Plate Recognition

License plate recognition occurs after the system detects vehicle movement. If the camera is not positioned to reliably detect vehicle movement, license plate recognition may also be impacted. Additional camera placement requirements can help optimize license plate capture.

For more information, see the Managing License Plate Recognition (LPR) article.

Helpful Links

• Managing Facial Recognition
• Managing License Plate Recognition (LPR)
• Artificial Intelligence on the R360
• AI Confidence Thresholds
• Configuring Activity Regions
• Optics and Object Distances for Analytics

Contact Support or Sales

Have more questions? Contact Rhombus Support at +1 (877) 746-6797 option 2 or support@rhombus.com.

Interested in learning more? Contact Rhombus Sales at +1 (877) 746-6797 option 1 or sales@rhombus.com.