ABSTRACT
This chapter presents the detailed analysis of detection, identification, and count of vehicles in an image, using convolutional neural networks CNN)-based YOLO (You Only Look Once) architecture. The proposed approach is applied in civil fields like automated toll revenue calculation and the calculation of expected strength of the road. The vehicles were detected, identified, and counted in real time based on the proposed architecture. The identification of any vehicle is often accomplished through sensing element gear. Examples of these sensing element gears include inductive circle finder, infrared symbol, and measuring device indicator. However, to chop the price, we tend to use YOLO architecture trained on the COCO (Common Objects in Context) dataset, and it’s capable of detecting Eighty Common Objects in context.
