Early Warning System for Safe Operation of Grain Silos based on Intelligent Video Analysis

Abstract:
China is a country with a large population. Ensuring the effective supply of grain is of great significance to ensuring the stable operation of the national economy. In order to improve the security capability of grain purchase, storage and supply, it is necessary to carry out intelligent upgrading and transformation of grain depots. At present, the identification of violations in the granary is mainly by installing sensors in key areas and manually reviewing video records. This method not only takes up a lot of resources, but also is unsatisfactory in real-time and accuracy. This paper proposes a video behavior recognition algorithm based on deep learning, combined with the analysis of the unsafe operation behavior of granary workers, to achieve effective, real-time recognition and reminders to ensure the safety of food and employees.