Real-time Prediction Program of Occupancy Changes in a Large Exhibition Hall using Deep Learning Approach

摘要

A method and device for real-time prediction of changes in the number of occupants in a large exhibition hall based on deep learning are presented. The deep learning-based real-time prediction method for occupancy changes in large exhibition halls proposed by the present invention divides the space to predict the occupants, pre-processes data about the occupants in the space collected through simulation, and deep processing of the pre-processed data. Creating time series data for learning learning, using the generated time series data to learn a deep learning model to predict the number of occupants for each divided area, and in the space collected in real time through server and socket communication and predicting the number of occupants in the space by inputting data about the occupants into the learned model.