Showing 1 ideas for tag "transit"

Mobility and Operations

Innovative Video Analytics & Edge Computing Systems for Traffic Volume and Incidents Detection trending idea

For the past five years, we have been working on video analytics R&D projects sponsored by NJDOT and NSF for traffic incident detection, traffic volume counting, traffic congestion management, and pedestrian detection. 

Since 2021, we have been exploring the possibility of field deployment of our smart traffic video analytics systems for NJDOT, and demonstration projects for pedestrian and vehicle detection, counting,... more »

Idea Submitted by Chengjun (CJ) Liu, Professor of Computer Science

List your Agency /Division / Bureau, County, City Univ. or Other New Jersey Institute of Technology, Newark, NJ 07102

Idea Champion - Who at NJDOT, NJ Transit or NJ Motor Vehicle? (if known) Wasif Mirza, Amanda Gendek

Why is this a priority and what are the benefits for the State/County/City?

The proposed innovative idea and advanced video analytics/edge computing technologies will revolutionize the current state of practice by applying innovative Artificial Intelligence (AI), Machine Learning (ML), Computer Vision (CV), Pattern Recognition (PR), Edge Computing (EC) to most efficiently analyze the real-time video data provided by the current available infrastructure such as video cameras. Such precious real-time data will be processed and analyzed in real-time and the results will be provided to authorities for real-time decision making in order to prevent crimes or tragedies before they happen. Thus property costs and even human lives may be saved.

Our innovative technologies achieve state-of-the-art performance, and improves upon traditional techniques such as Radar by nearly an order of magnitude improvement for traffic volume counting. Our advanced video analytics technologies, which tackles numerous challenging factors, such as low quality video, low bit rate video, adverse illumination and weather conditions, improve upon other methods by large margins.


2 votes