Automatic Billing System using Resnet and Image AI

Abstract

In this age of automation, industries are trying to modernize their infrastructure by installing robotic and automated machineries in their stores and workshops. This is done in order to reduce the cost of manufacturing a product, speeding up the process and making the industry less dependent on manual labor. In retail stores and shopping malls, customers have to wait in line at the billing counter for the receipt of the items which they brought. This is due to the inefficiency of the human worker who enters the details of the product and makes bill. This is a time consuming process. To make it more efficient we propose an automated billing system which can calculate the total billing amount of the items brought without any help of cloud serves and internet. We use the concept of edge computing to make our system less dependent on internet and more dependent on local machines. We use the techniques of image classification like residual network and ImageAI to identify the product with different dimensions and color.