Crowd Surveillance System

Crowd Surveillance System

Gayatri Shinde, Pratibha Adhav, Lokesh Malkani, Pooja Pedgaonkar | Prof.Ms.Ekta Choudhari

Abstract

A huge number of studies related to features for improving the accuracy of object detection systems using the Convolutional Neural Network (CNN). The majority of projects/research is based on free parking space which is a slow model and which is based on the CNN model. In object detection technique monitoring the objects using by using the traditional manual technique, the video surveillance technique is feasible to use when there is a vast number of objects to monitor. To improve real-time problems whose accuracy should be fast and can be used in day-to-day life. To enhance the accuracy and speed of the CNN algorithm we can add an extra to the model which is known as YOLO (You Only Look Once). It enables the system that can recognize objects in the provided frame. Real-time object detector operation requires Graphics Processing Unit (GPU) for object detection [1].