Year: 2025
Type: article
Abstract: - This project aims to transform the conventional shopping experience by introducing a smart, handsfree checkout system. Using computer vision and machine learning, it simplifies the billing process, reduces waiting times, and improves overall customer satisfaction. In typical retail settings, long queues and manual billing often lead to delays and errors, frustrating customers. To tackle this, the system allows users to simply take a photo of their purchased items using their mobile phone. This image is then uploaded to a user-friendly web application built with Flask. The core of the system is the YOLOv10 object detection model, which is trained on a custom dataset containing 4–5 categories of common shopping items. Once the image is uploaded, the model identifies each item in the photo, retrieves the price from a backend database, and instantly generates the total bill. The platform offers a seamless and intuitive experience, removing the need for manual scanning or cashier involvement. While the payment process is currently simulated through a mock success page, it demonstrates the potential for full automation in future versions. This prototype showcases how emerging technologies can create a smoother, more efficient, and handsfree retail checkout experience. Key Words: Flask Web Application, Machine Learning, Real-Time Detection, Prototype Development, Automated Billing, Payment Simulation
DOI: 10.55041/ijsrem49021
Author: Prof. S. M. Shelke
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4410857635