A Novel Survey on ML based Vehicle Detection for Dynamic Traffic Control

摘要

In recent decades, object detection has played animportant role in real-time applications. The threat of accidents and criminality was mounting, and the population was growing quickly as a result, covering technology is improving on a daily basis classifying items is a simple activity for humans but it is a difficult assignment for robots owing to the variety of elements, where colourful models have been used to break up this sensitive process. There are several methods for depicting multiple things this study examines several models that have been preliminarily imposed to describe numerous items in real time. This becomes a challenging task as it takes a long time to reuse an accurate object recognition. Hence, developing a novel architecture to perform genuine object detection is highly required. This study provides a method for detecting numerous items and discusses about different machine learning models, including R-CNN, SSD, and YOLO. It also indicates that several designs were utilised to build A frame for holding generally realistic models for detection, recognition, and identification.