Environmental Sustainability Through Optimal Energy Consumption Using IoT-Based Edge-Computing and Image Processing

This chapter explored various techniques and modelling methodologies designed to estimate occupancy level in indoor environments. It also introduces an innovative image-based occupancy detection system that leverages edge computing and machine vision to accurately detect and classify occupants and other objects in an indoor environment which requires a certain thermal comfort level. By enabling real-time adjustments to HVAC operations based on actual occupancy, it can significantly reduce unnecessary energy consumption in unoccupied areas, thus improving overall energy management. The integration of edge computing allows for local data processing, which not only minimizes the computational load on centralized servers but also addresses privacy concerns by reducing the need for external data transmission. This is particularly important in environments where sensitive information about occupants may be captured. A case study is presented in the end to demonstrate and examines the performance of several object detection models in the context of academic office occupancy detection.