International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 04 | Apr 2024
p-ISSN: 2395-0072
www.irjet.net
AutoBilling System Anas Usmani1, Abhinav Pandey2, Pratham Solanki3, Rahul Yadav4, Zainab Mizwan5 Shree L R Tiwari College of Engineering, Mira Road ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - The AutoBilling system represents a significant
level features. The review covers various deep learning architectures, modifications, and specific detection tasks, with experimental analyses and future research directions provided [1]. The paper proposes a Hybrid PCA-SIFT-FREAK algorithm for 3D object recognition in an automated billing system. It balances accuracy and memory usage, offering advantages over traditional methods. Comparative study shows improved performance, with real-time application potential, achieving high recognition rates while reducing computation time and memory requirements. [2].
leap forward in the retail industry, harnessing the power of machine learning and computer vision to redefine the shopping experience. Traditionally, barcode scanning has been the go-to method for product identification and billing, but it often proves time-consuming and labor-intensive for both customers and staff. By leveraging cutting-edge technology, AutoBilling eliminates the need for manual barcode scanning, offering a seamless and contactless alternative. Using sophisticated algorithms, the system swiftly detects and identifies products as they are placed on the counter, significantly reducing wait times at checkout. This not only enhances efficiency but also minimizes human interaction, a crucial consideration in today's health-conscious climate.
This paper outlines machine learning's dual focus on constructing systems that improve through experience and understanding the statistical laws governing learning processes. It highlights machine learning's evolution from a laboratory concept to a practical technology widely used in commercial applications, spanning various fields from AI to empirical sciences. [3]. This paper proposes a smart unstaffed retail shop scheme utilizing image processing with Python to enhance the shopping experience. By training an end-to-end classification model on a dataset of images containing various stock keeping units (SKUs), the system achieves accurate SKU recognition and counting, addressing limitations of traditional unmanned containers. [4].
Key Words: AutoBilling, retail industry, machine learning, computer vision, shopping experience, cutting-edge technology, contactless.
1.INTRODUCTION AI-Driven Billing System is an AI-powered autonomous checkout system for retail stores that utilizes computer vision and machine learning to offer a faster and more efficient shopping experience. The system is designed to minimize human interaction in the store to keep shoppers and employees safe during the pandemic. The system works by using computer vision algorithms and machine learning models to visually detect and identify items placed on the counter-top.
This paper introduces MobileNets, a class of efficient models designed for mobile and embedded vision tasks. MobileNets utilize depth-wise separable convolutions to construct lightweight deep neural networks. The authors propose two global hyperparameters for balancing latency and accuracy, enabling model customization based on application constraints. Extensive experiments demonstrate MobileNets' strong performance across various applications including ImageNet classification, object detection, fine-grain classification, face attributes, and geo-localization. [5]. Considering the time constraints that it takes to store memory; OMCL is proposed with a Phase shift mechanism. Just doing this increases the performance by 86.1% but it also hampers the life of the system by a miniscule 3.4% [6].
The system is equipped with cameras that capture high resolution images of the products, and a deep learning model processes these images in real-time to recognize the products accurately. Once the products are recognized, the system automatically adds them to the virtual cart and generates an itemized bill. In addition to visual recognition, AI-Driven Billing System also utilizes a weight sensor to measure the weight of the items placed on the counter-top. The system uses this information to ensure the accuracy of the billing process and prevent any discrepancies. One of the most significant advantages of AI Driven Billing System is its ability to generate an instant bill.
2.RELATED WORK This paper reviews the evolution of object detection from traditional methods to deep learning frameworks. It discusses the limitations of traditional approaches and highlights the advantages of deep learning in learning high-
© 2024, IRJET
|
Impact Factor value: 8.226
|
ISO 9001:2008 Certified Journal
|
Page 1172