5. Object Detection and Tracking Time
Ensure vehicles remain in camera view for at least 2 seconds by adjusting FOV based on speed; optimise frame rate and positioning for accurate detection.
Understanding the minimum time a vehicle must remain within the camera's field of view (FOV) is critical for accurate detection, tracking, and counting by the XVision Video Analytics Platform. This section emphasises the importance of FOV configuration and the required object presence time to ensure reliable system performance across various traffic conditions.
Summary of Key Points
Minimum Object Presence Time:
Vehicles must remain within the camera's FOV for at least 2 seconds for accurate detection and tracking.
For tripwire based measurements the object must be in the scene at least 1 seconds before the line and 1 seconds after the line. This means the object must be in the scene for at least 2 full seconds for accurate counting can occur
Field of View (FOV) Adjustments:
Calculate Required FOV Length:
Use the vehicle speed and minimum presence time to determine the necessary FOV length.
Extend FOV:
Adjust camera positioning, lens selection, or deploy additional cameras to cover the required area.
Camera Settings Adjustments:
Frame Rate:
Increase frame rate to capture more frames of fast-moving vehicles.
Image Quality:
Optimise noise reduction and compression settings to maintain image clarity.
Optimising Detection:
Lens Selection and Camera Placement:
Choose appropriate lenses and positions to maximise the effective FOV.
Regular Testing:
Continuously assess and adjust the system to ensure reliable performance.
Minimum Object Presence Time
Required Time in View: At least 2 seconds for the system to accurately detect a vehicle, acquire a lock, and begin tracking it through the scene.
Importance of Minimum Object Presence Time
Detection Process
Frame Analysis:
The system analyses multiple consecutive frames to confirm the presence of a vehicle.
Processing several frames reduces the likelihood of false positives from transient objects.
Motion Tracking:
Consistent detection over time allows the system to establish the vehicle's trajectory.
Essential for accurate counting and distinguishing between vehicles, especially in heavy traffic.
Implications for Field of View (FOV)
Calculating Required FOV Length Based on Vehicle Speed:
Formula
FOV Length (meters) = Vehicle Speed (meters/second) * Minimum Object Presence Time (seconds)
Example Calculations:
At 60 km/h (16.67 m/s):
16.67m/s * 2s = 33.34 meters
Required FOV Length: At least 33.34 meters along the vehicle's path.
At 100 km/h (27.78 m/s):
27.78m/s * 2s = 55.56 meters
Required FOV Length: At least 55.56 meters.
Adjustments for Higher Vehicle Speeds:
Extend FOV:
Adjust the camera's FOV to cover a longer section of the road to accommodate faster vehicles.
Use lenses with a wider horizontal field of view or reposition the camera further from the road.
Use Multiple Cameras:
Deploy additional cameras with overlapping FOVs to maintain continuous tracking.
Ensures vehicles remain within the monitored area for the necessary duration.
Camera Placement:
Position cameras strategically to maximize the length of the vehicle's path within the FOV.
Angling the camera along the direction of traffic flow can increase the effective monitoring distance.
Adjusting Camera Settings
While shutter speed adjustments are excluded, other camera settings can be optimised to enhance detection without altering shutter speed.
Frame Rate
Importance:
Capturing Motion Detail:
Higher frame rates provide more frames per second, increasing the likelihood of capturing clear images of fast-moving vehicles.
Reduces motion blur and ensures the system has sufficient data for accurate detection.
Recommendations:
Standard Applications:
Set frame rate to 15-25 fps for typical traffic conditions.
High-Speed Traffic Areas:
Consider increasing the frame rate to 50-60 fps for areas with very fast-moving vehicles.
Note that higher frame rates may not be achievable in all Xvision processing units. Please refer to Minnovation for specific advice.
Image Quality Settings
Noise Reduction:
Purpose:
Enhances image clarity by reducing noise, especially in low-light conditions.
Implementation:
Enable 3D noise reduction features within the camera settings.
Adjust the level of noise reduction to balance image sharpness and clarity.
Compression Settings:
Balancing Quality and Bandwidth:
Use efficient video codecs like H.265 to maintain high image quality while conserving bandwidth.
Avoid excessive compression that could degrade image detail and impact detection accuracy.
Optimising Field of View for Various Conditions
Lens Selection
Wide-Angle Lenses:
Provide a broader FOV, allowing vehicles to remain in view longer as they pass through the scene.
Suitable for areas where vehicles move at higher speeds and require extended monitoring distance.
Telephoto Lenses:
Offer a narrower FOV with greater magnification.
May be appropriate for monitoring specific points where vehicles slow down, such as intersections or toll booths.
Camera Positioning
Height and Angle Adjustments:
Elevate the camera or adjust its angle to increase the length of the roadway covered within the FOV.
Ensure that the camera's perspective does not introduce excessive distortion.
Distance from Road:
Position cameras further from the road to naturally extend the FOV along the vehicle's path.
Balance the distance to maintain sufficient image resolution for accurate detection.
Regular Testing and Calibration
Performance Monitoring:
Conduct field tests to verify that vehicles remain within the FOV for the required minimum presence time.
Adjust camera positioning and settings based on test results.
Seasonal and Environmental Considerations:
Re-evaluate FOV and object presence time as environmental conditions change (e.g., foliage growth, construction).
Ensure that changes in the surroundings do not reduce the effective monitoring area.
By focusing on the minimum object presence time and optimising the field of view accordingly, you can significantly enhance the accuracy and reliability of your vehicle monitoring system. Proper configuration of camera placement and settings ensures that the XVision Video Analytics Platform performs optimally across various traffic speeds and environmental conditions.
Last updated
