AI (Artificial Intelligence) intelligent vending cabinet based on dynamic visual recognition

Abstract

The invention discloses an AI (Artificial Intelligence) intelligent vending cabinet based on dynamic visual recognition. The vending cabinet is communicatively connected with a cloud server. The vending cabinet comprises a cabinet body provided with at least one shelf, a camera disposed on the cabinet body, and a control module connected with the camera and a wireless transmission module; the cloud server includes a dynamic visual recognition algorithm processing module and a storage space; the camera shoots the whole process that an object departs away the cabinet body into a video and the video is uploaded to the cloud server through the wireless transmission module; video processing calculation recognition is performed by the dynamic visual recognition algorithm; and the recognized result is sent to a user payment terminal through a network for fee deduction. According to the AI intelligent vending cabinet based on dynamic visual recognition disclosed by the invention, the number ofhardware components of the vending cabinet can be decreased, the assembly of the hardware is simplified, the superposition placement of multiple commodities in the vending cabinet is implemented, thus the cost of the whole machine is reduced, the space utilization rate of the vending cabinet is improved, the unmanned retail is implemented and the retail labor cost and the site rental cost are greatly reduced.

Description

technical field [0001] The invention relates to the technical field of intelligent vending cabinets, in particular to an AI intelligent vending cabinet based on dynamic visual recognition. Background technique [0002] At present, there are mainly two identification methods for smart vending cabinets, one is based on traditional RFID technology, and the other is based on visual recognition technology. The disadvantages of traditional RFID technology are high equipment and operating costs, and RFID technology is greatly affected by liquid and metal, and there is an insurmountable technical bottleneck. Visual recognition technology also includes static recognition and dynamic recognition. Static recognition uses photos to compare and identify products, and dynamic recognition uses video to compare and identify products. Among them, static recognition needs to install 1-2 cameras next to each shelf in the vending cabinet, and the cameras need to be at a higher height from the ...

AI Technical Summary

Problems solved by technology

Among them, static recognition needs to install 1-2 cameras next to each shelf in the vending cabinet, and the camera needs to be at a higher height from the lower shelf, otherwise it is difficult to capture the whole picture. Therefore, the space utilization rate of static recognition is low

In addition, the vending cabinets for static identification have higher requirements on the placement of products, the products cannot be stacked, and it is difficult to identify products with the same appearance, etc.

Images