A Genetic Algorithm-Based System for Automatic Product Recognition and Optimization in Supermarkets

With the continuous development of intelligent retail, traditional supermarkets encounter issues such as low efficiency and excessive manual intervention in the processes of product recognition, weighing, and checkout. To address these challenges, this paper proposes an automatic product recognition system for supermarkets based on genetic algorithms. The system integrates image recognition and intelligent weighing technologies, aiming to enhance supermarket operational efficiency and improve customer shopping experience. The system employs the K210 chip and convolutional neural networks (CNNs) for real-time product image acquisition and recognition. It utilizes genetic algorithms to optimize recognition accuracy, reducing misidentifications and improving precision. Furthermore, by incorporating a weighing module, the system can automatically identify products and calculate prices, enabling self-checkout functionality. Experimental results demonstrate that the system operates stably across various environments, with errors within an acceptable range. Compared to traditional manual operations, the system significantly improves checkout efficiency and provides a viable technological solution for intelligent retail. The findings of this study offer new insights and technical support for the future development of smart supermarkets, with broad application prospects.