Rancang Bangun Sistem Monitoring Aktivitas Pelanggan berbasis Citra Menggunakan Optical Flow

Abstract

This article aims to detect visitors using the optical flow method approach. It is based on transaction activities that are not freed from human life especially on the basic needs of making the spending trend in physical stores has begun to be crushed by the competition of online stores. Several factors are the inability of stores employees in serving the buyer up to the process of non-instant payment makes the physical store threatened that can imply on physical assets such as display cabinets, cash machines, to buildings. Based on the problem, the research was carried out by creating a system that can detect the activity of visitors on the store. The research was done by processing three video inputs in different cases. With the application of pre-processing processes and the implementation of optical flow methods, results were obtained that the system was able to detect the movement of visitors well.