Near Miss Event Detection System


Grid- based spatial autocorrelation analytics


Key features of our system

High-Resolution Analysis

High Robustness

High Accuracy

Pixel Level Video Stabilization

Car Pose Estimation

Lost and Missing Vehicles Re-matching

Semantic Segmentation

Fusion of Deep Learning and Traditional Method

System Overview

This Near Miss Event Detection System (NMEDS) is developed by the UCF SST team to conduct traffic analysis using video data collected from roadside cameras. The framework of NMEDS combined the Mask-RCNN bounding box and Occlusion-Net detection algorithms to reconstruct road users’ key points in a 3D view. The following are some examples of traffic analysis that could be done using the system:

System Demo

Hours Testing
Sample Size

Our Performance

Accuracy – 95%+
Expected Average Overlap (EAO) – 95%+
Robustness – 85%+

OUR Example

What we’ve done for safety