People Counting System for Smart Energy Consumption and Mob Management

Abstract

The new rule for the future is going to be, “Anything that can be connected, will be connected”. The system in this paper is proposed in order to take a step towards a connected and smarter future, and also to foster an efficient way for management of crowd and energy consumption. This system is built using various trends driving the future of information technology such as Raspberry pi, Image processing, MQTT publish-subscribe protocol, ESP8266 NodeMCU, and Node-RED. Raspberry pi camera module is used for capturing images at regular intervals of time; the captured image is processed by Raspberry pi using OpenCV and Haar cascade classifier which detects human heads to generate a count of the detected people. This count is published to Node-RED by Raspberry Pi which is subscribed by the ESP8266 NodeMCU to control the electronic devices according to the subscribed count, rather according to the presence or absence of people. MongoDB database connectivity is given to Node-RED wherein statistics about crowd density and distribution at a particular time and area are obtained.