Model Design Research Plan for Warehouse Barcode Image Recognition in Smart Systems

This article presents a comprehensive plan for creating a research model of a neural network for barcode recognition in a warehouse environment. The introduced new methodology enables the development of an efficient model with high accuracy for barcode recognition and decoding. In the proposed methodology, we consider current approaches and present a research plan focusing on convolutional neural networks and the Roboflow tool. The new methodology has been divided into several steps, including data collection, dataset compilation, selection of a suitable neural network model, network training, validation, and optimization, and finally, testing the model on a test dataset.