Video-analysis-based antitheft monitoring method for physical store

Abstract

The invention discloses a video-analysis-based antitheft monitoring method for a physical store. The method comprises the following steps of: tracking the motion trail of a customer in the physical store in a video analysis way, identifying the hand motion and upper limb motion of the customer at a specified place of the physical store, and judging whether the customer conducts suspected theft or not by combining commodity changes before and after the customer leaves the specified place; and when the customer is determined to conduct the theft, continuously tracking the customer, and transmitting a prompt to service staff or security staff in the store when the customer tends to leave the physical store. The suspected theft information of the customer when the customer browses commodities in the physical store can be quickly and accurately captured, and theft judgment accuracy is improved through continuous tracking. Therefore, economic loss caused by the theft of the customer to the physical store is greatly reduced on the premise of substantially no increase of labor cost, and the economic benefits of the physical store are improved.

Description

technical field [0001] The invention relates to an anti-theft monitoring method, in particular to an anti-theft monitoring method for a physical store realized based on video image analysis technology, and belongs to the technical field of intelligent video analysis. Background technique [0002] Brick and mortar stores have long dominated the retail industry. A large number of customers visit typical physical stores every day, such as clothing stores, gift stores, accessory stores, supermarkets, etc., among which there will be a small proportion of thefts, but the absolute number and the amount involved are very alarming. These thefts mainly target products that are easy to hide in the body, such as foldable clothes, small gifts, and small packaged foods. For this kind of theft, if you still rely on the in-store service personnel or security personnel to carry out manual monitoring, the corresponding human and material resources will be very huge. [0003] In order to sol...

AI Technical Summary

Problems solved by technology

[0006] However, the existing video surveillance system still needs to manually watch the screen all the time, and cannot actively prompt the suspected theft, let alone send the necessary prompts when the suspected theft occurs—especially when the suspect leaves the scene

Images