We have taken a safety-first approach at bike detection in the box, particularly since short extensions for fast accelerating or moving objects have virtually no impact on efficiency. This is especially true when compared to traditional stop-line bike discrimination that can create long-lasting minimum greens for a FP detection. With that in mind, we err on the side of caution in the box and allow “uncertain” objects to be classified as bicyclists but require more certainty to record a bike versus vehicle count. Pedestrians in or near the box, such as crosswalks directly adjacent to the drawn box, can also trigger FP detections in the box. GRIDSMART will continue to improve on this front, reducing the FP rate while preserving accurate detection.
What are the some causes of false positive (FP) bike detections in the box? Print
Modified on: Wed, 1 May, 2019 at 5:00 PM
Did you find it helpful?Send feedback
Sorry we couldn't be helpful. Help us improve this article with your feedback.