ETMS: Efficient Traffic Management System for Congestion Detection and Alert using HAAR Cascade

Abstract

Rapid social development has resulted in the emergence of a new major societal issue: urban traffic congestion, which many cities must address. In addition to making  it more difficult for people to get around town, traffic jams are a major source of the city's pollution crisis. In order to address the problems of automobile exhaust pollution and congestion, this paper uses the system dynamics approach to develop a model to study the urban traffic congestion system from the perspectives of trucks,private cars, bikes and public transportation. This project proposes a system for detecting vehicles and sending alerts when traffic levels rise to dangerous levels using Haar Cascade and Fuzzy Cognitive Maps (FCP). The proposed system uses Haar Cascade to detect moving vehicles, which are then classified using FCP. The system can make decisions based on partial or ambiguous information by utilising FCP, a soft computing technique, which allows it to learn from past actions. An algorithm for estimating traffic density is also used by the system to pinpoint active areas. In congested areas, the system will alert the driver if it anticipates a collision with another vehicle and also Experiments show that the proposed system is able to accurately detect vehicles and provide timely alerts to the driver, drastically lowering the probability of accidents occurring in heavily travelled areas.

The importance of introducing such a system cannot be overstated in today's transportation system. It's a big deal for the future of intelligent urban planning and traffic control. Congestion relief, cleaner air, and increased security are just some of the long-term benefits that justify the high initial investment. To add, this system is adaptable to suburban and rural areas, which can also experience traffic congestion issues.