The use of computer vision technologies to monitor the availability of bakery products on shelves in real time

Abstract

The rapid development and introduction of innovative technologies, in particular computer vision, opens up new prospects for optimizing processes in various industries, including the baking industry in Russia. This study is devoted to the study of the possibilities of using computer vision systems to control the availability of bakery products on store shelves in real time. The work analyzes existing solutions based on computer vision technologies and evaluates their effectiveness in the context of the Russian bakery market. Research materials and methods include the analysis of scientific publications, patents and practical cases related to the use of computer vision in retail, as well as conducting a series of experiments in real stores. During the experiments, video surveillance systems with a resolution of 1080p and a frame rate of 30 fps were used, as well as specialized software for image processing and analysis based on deep learning algorithms and convolutional neural Networks (CNN). The results of the study demonstrate that the introduction of computer vision technologies makes it possible to increase the effectiveness of monitoring the availability of bakery products on store shelves by 25-30% compared with traditional methods based on manual monitoring. The computer vision system is capable of determining the presence or absence of goods on the shelf in real time with up to 95% accuracy, as well as identifying specific types of bakery products. In addition, the use of computer vision can reduce staff labor costs by 15-20% and reduce sales losses associated with the absence of goods on the shelves by an average of 10-12%.