الفهرس 
Chapter 1.Only 14 pages are availabe for public view
 |  | from | 104 |  |  |
| | | from | 104 | | |
Abstract
Car congestion is an urgent issue for many people. Accidents, traffic lights, rapid accelerations, deceleration, and hesitation of drivers, and low-carrying capacity roads without bridges can cause car congestion. Increasing the road width and establishing rounds and bridges are solutions for car congestion, but these are expensive. Traffic light recognition (TLR) decreases accidents and congestion caused by traffic lights (TLs). In a semi-automatic annotation for traffic light detection, data under harsh conditions were not collected and tracking was not supported. Integrated channel feature tracking (ICFT) combines detection and tracking; however, sharing information with neighbors is not supported.Vehicular ad-hoc networks (VANETs) address a steadily expanding demand, particularly for public emergency applications, because the real-time localization of destination vehicles is important for determining the route to deliver messages. Existing location administration services in VANETs are classified into flooding-based, flat-quorum-based, and geographic-based location services. However, existing localization techniques suffer from network disconnection and overloading because of the VANET topology changes. A traffic light-inspired location service (TLILS) is proposed to manage localization inspired by traffic lights. Roadside Units (RSUs) are used as location servers by the proposed optimized localization service. RSUs with maximum traffic weight metrics were chosen based on the speed of vehicles, network connection time, and density of neighboring vehicles. The proposed TLILS outperforms both Name-ID Hybrid Routing (NIHR) and Zoom-Out Geographic Location Service (ZGLS) in terms of packet delivery ratio (PDR) and delay. Vehicular ad-hoc networks (VANETs) have been used for VANET traffic light recognition (VTLR). Sharing information and tracking the TL status, time to change, and advised speeds for vehicles were supported. Measurements show that VTLR outperforms semi-automatic annotation and ICFT in terms of delay, success ratio, and the number of detections per second.